Protecting Reliance - Victor P. Goldberg PDF

GOLDBERG, Victor P. "Protecting Reliance". En: Columbia Law Review, Vol. 114, No. 4 (MAY 2014), pp.
1033-1081
ESSAY
PROTECTING RELIANCE
Victor P. Goldberg *
Reliance plays a central role in contract law and scholarship. One

party relies on the others promised performance, its statements, or its
anticipated entry into a formal agreement. Saying that reliance is
important, however, says nothing about what we should do about it.
The focus of this Essay is on the many ways that parties choose to protect
reliance. The relationship between what parties do and what contract
doctrine cares about is tenuous at best. Contract performance takes
place over time, and the nature of the parties future obligations can be
deferred to take into account changing circumstances. Reliance matters
in this context since one or both of the parties might want to rely on the
continuity of the arrangement; but they might also want the flexibility to
adapt as new information becomes available. If one party has the discre-
tion to react (terminating or adjusting quantity, for example), the other
party can confront the decisionmaker with a price reflecting its reliance.
That price need bear no relationship to the formal remedies of contract
law. The Essay also considers other roles of reliance, including how it is
used to determine compensation following excused performance, to trig-
ger the irrevocability of an offer, and to validate information in a parti-
cular type of complex transactiona corporate acquisition.
INTRODUCTION
Reliance-talk permeates contract law and scholarship. One party
relies on the others promised performance, its statements, or its anti-
cipated entry into a formal agreement. Often reliance is paired with
justice (or, more precisely, avoiding injustice) for defining obligations or
reckoning compensation. It is the centerpiece of the Fuller-Perdue tri-
umvirate of interests to protect.1 Reliance, they argued persuasively, is
tremendously important.2 And many scholars were persuaded.3 It is one
*. I would like to thank Ron Gilson, Mark Lawson, Bob Scott, George Triantis, and
the participants in a workshop at Pepperdine University School of Law for their helpful
comments, and Ni Qian and Carson Zhou for their research assistance.
1. See L.L. Fuller & William R. Perdue, Jr., The Reliance Interest in Contract
Damages: 1, 46 Yale L.J. 52, 5356, 7175 (1936) [hereinafter Fuller & Perdue, Reliance 1]
(identifying restitution, reliance, and expectation interests as three principal purposes in
awarding contract damages and noting restitution and expectation interests both converge
with reliance interest).
2. Id.
1033
1034 COLUMBIA LAW REVIEW [Vol. 114:1033
of the most cited articles in contract law. 4 Saying that reliance is

important, however, says nothing about what we should do about it. The
leap from that proposition to implications for doctrinea leap made by
Fuller-Perdue and many followersdoes not follow.5 The Fuller-Perdue
framework does nothing to resolve the Goldilocks problem: Is the pro-
tection of reliance too much, too little, or just right?
Too often court decisions or scholarship put forward arguments of
this form: People want to be able to rely upon X, and if the law does not
take that reliance into account, then something bad (inefficiency for
some, injustice for others) will occur. Principles of contract doctrine can
then be invoked, rejiggered, or even amended, so that the reliance is
properly accounted for. The discussion, I believe, typically fails to ask the
most basic questions: What do parties do and why do they do it? The
focus of this Essay will be on these contract-design questions. Under-
standing how and why parties deal with reliance questions will provide
insights into both contract doctrine and interpretation.
Contract law allows parties, within constraints, to set their own rules.
It provides a set of default rules, and if the parties do not like them, they
can change them. Some of the rules, however, are mandatory, and others
are, to varying degrees, sticky. That stickiness is in part due to the costs of
tailoring a transaction. For example, in many business-to-business (B2B)
transactions, including million-dollar deals, the documentation consists
of conflicting language on the back of two printed forms. Apparently in
such cases the parties would rather rely on the default rule for resolving
the conflicting language than spell out their obligations in a negotiated
document.6
A more subtle source of stickiness is the beliefs of judges, scholars,
and lawyers about the nature of contracts and contractual duties. The
notion that victims of a breach of contract should be made whole is a
3. P.S. Atiyah, Fuller and the Theory of Contract, in Essays on Contract 73, 73 (1986)
(The Reliance Interest in Contract Damages[] has probably been the most influential
single article in the entire history of modern contract scholarship, at any rate in the
common law world. (footnote omitted)); P. Linzer, A Contracts Anthology 285 (2d ed.
1989) (The Fuller and Perdue article is probably the best-known article in the contracts
literature . . . .); Todd D. Rakoff, Fuller and Perdues The Reliance Interest as a Work of
Legal Scholarship, 1991 Wis. L. Rev. 203, 204 ([I]ts substantive propositions are still
considered worthy of debate; its themes provoke new studies of great merit; and its
concepts are incorporated into the latest Restatement of Contracts. (footnotes omitted)).
4. Fred R. Shapiro & Michelle Pearse, The Most-Cited Law Review Articles of All
Time, 110 Mich. L. Rev. 1483, 1490 (2012).
5. For criticisms of the Fuller-Perdue argument for awarding reliance damages, see
Richard Craswell, Against Fuller and Perdue, 67 U. Chi. L. Rev. 99, 111 (2000), and
Michael B. Kelly, The Phantom Reliance Interest in Contract Damages, 1992 Wis. L. Rev.
1755, 1758.
6. On the costs and benefits of providing specific terms at the front end versus having
a third party determine the content of the agreement at the back end, see Robert E. Scott
& George G. Triantis, Anticipating Litigation in Contract Design, 115 Yale L.J. 814, 82239
(2006) [hereinafter Scott & Triantis, Anticipating Litigation].
2014] PROTECTING RELIANCE 1035
powerful one that constrains the ability of parties to structure their rela-
tionship.7 The Uniform Commercial Code (U.C.C.) and Restatement
(Second) of Contracts are peppered with notions of good faith and pre-
vention of injustice.8 There is a vast literature on the philosophy of
contract-law morality.9 I do not intend to engage with it directly. I do
hope that by focusing on the question of contract design, I can nudge
contractual ethos and contract doctrine in a direction that is more
congenial to the needs of the parties.
My concern, I must emphasize, is with the contracts of sophisticated
parties. I am not concerned with consumer contracts or agreements
between amateursfor example, an uncles promise of cash to pay for a
nephews car, which was featured in the debate over the adoption of
section 90. 10 There can, of course, be some dispute over whether a
particular contract falls in that category. For my purposes, the pertinent
category is defined by agreements for which both parties could be
expected to have access to counsel.11
Reliance is implicated in a number of doctrinal areas. Instead of
starting with doctrine and working out, this Essay begins with the design
problems and works backward to doctrinal considerations. The first
design problem concerns the tradeoff between reliance and flexibility.
7. Scott and Triantis stress both the importance of the compensation principle in
contract law and its dubious foundations. Robert E. Scott & George G. Triantis, Embedded
Options and the Case Against Compensation in Contract Law, 104 Colum. L. Rev. 1428,
142829 (2004) [hereinafter Scott & Triantis, Embedded Options].
8. See, e.g., U.C.C. 2-508 cmt. 2 (2012) ([U.C.C. 2-508(2)] seeks to avoid
injustice to the seller by reason of a surprise rejection by the buyer.); id. 2-610 cmt. 3
(noting party may sue for damages resulting from anticipatory repudiation when material
inconvenience or injustice will result if the aggrieved party is forced to wait and receive an
ultimate tender minus the part or aspect repudiated); Restatement (Second) of Contracts
90(1) (1981) (A promise which the promisor should reasonably expect to induce action
or forbearance on the part of the promisee or a third person and which does induce such
action or forbearance is binding if injustice can be avoided only by enforcement of the
promise.); id. 272(2) (permitting courts to grant relief on such terms as justice
requires including protection of the parties reliance interests).
9. See, e.g., Charles Fried, Contract as Promise 1417 (1981) (noting contract
obligation is special case of general moral duty to keep promises premised on mutual
trust); T.M. Scanlon, What We Owe to Each Other 295309 (1998) (arguing obligations to
keep promises arise from principle of fidelity); Richard Craswell, Promises and Prices, 45
Suffolk U. L. Rev. 735, 76667 (2012) (discussing moralitys role in shaping contract laws
default rules). See generally Symposium, Contract as Promise at 30: The Future of Contract
Theory, 45 Suffolk U. L. Rev. 601 (2012) (providing substantial commentary on whether
contract law is or ought to be grounded in moral duty).
10 . Gerald Griffin Reidy, Note, Definite and Substantial Reliance: Remedying
Injustice Under Section 90, 67 Fordham L. Rev. 1217, 1222 (1998) (noting framers
intended section 90 for enforcing relied-upon promises in purely donative setting, rather
than in commercial contexts).
11. Included in this category is the battle of the forms, alluded to above, even though
its essential feature is that the parties are not expected to involve counsel in the actual
transaction.
Because contracting parties need to adapt their relationship as circum-

stances change, they can choose to allocate discretion to one of the
parties. That discretion can be constrained to take into account the
counterpartys reliance. The discretion could take the extreme form of
termination or milder forms, such as granting one party the right to vary
quantity. Part I considers both the option to abandon and quantity
adjustment.
If one party relies on the other party to perform competently and
the counterparty fails to do so, the consequences can be severe. If I ship a
package and it fails to arrive, I, or the intended recipient, might suffer
consequential damages. Since Hadley v. Baxendale was decided, 12 the
aggrieved partys ability to obtain compensation for its resultant losses
has been limited. Part II considers the rationale for restricting the recov-
ery of consequential damages, with a particular emphasis on the role of
reasonable reliance.
When performance of a contract is excused (by impossibility or
related doctrines), the parties are left in the middle. A buyer might have
already paid thousands of dollars for goods that it will not receive
because the seller was excused. The seller might have spent substantial
sums in reliance upon the contract before the occurrence of an event
that excused performance. Must the seller disgorge the payments it has
received? Should the buyer compensate the seller for expenses made in
reliance upon the contract? The Restatement (Second) would have the
seller disgorge but also compensate it for reliance costs to avoid
injustice.13 English law yields a similar result.14 This is, I suggest in Part
III, a doctrine untethered by reality. Parties routinely design around the
doctrine.
Reliance, coupled with the prevention of injustice, shows up in other
corners of contract doctrine as well. It has been the basis for finding a
promise enforceable (section 90) and an offer irrevocable (section
87(2)).15 The rise of promissory estoppel and its practical significance (or
lack thereof) have been chronicled elsewhere, and I have little to add to
that. 16 The original source of the irrevocability argument is Justice
12. (1854) 156 Eng. Rep. 145 (Ex.); 9 Ex. 341.

13. Restatement (Second) of Contracts 272, 377.
14. See Law Reform (Frustrated Contracts) Act, 1943, 6 & 7 Geo. 6, c. 40, 1 (Eng.)
[hereinafter Frustrated Contracts Act] (providing seller must generally disgorge payments
received but allowing seller compensation for reasonable reliance). The Frustrated
Contracts Act was adopted shortly after Fibrosa Spolka Akcyjna v. Fairbairn Lawson Combe
Barbour, Ltd., [1943] A.C. 32 (H.L.) 4950 (appeal taken from Eng.), where the court
required the seller to return the prepayment but denied compensation to the seller for
costs it might have incurred in reliance on the contract. See infra notes 184193 and
accompanying text.
15. Restatement (Second) of Contracts 87(2), 90.
16. See, e.g., Stanley D. Henderson, Promissory Estoppel and Traditional Contract
Doctrine, 78 Yale L.J. 343, 344 (1969) (examining relationship between promissory
estoppel and familiar rules of contract formation); Avery Katz, When Should an Offer
Traynors famous decision in Drennan v. Star Paving Co.,17 which involved

an offer by a subcontractor to a general contractor. The practical signifi-
cance in contexts other than contracts between general contractors and
subcontractors in public projects is minimal. In the few cases in which
the issue has been litigated, the courts have shown little consistency in
identifying the type of reliance that would trigger irrevocability.18 Part IV
considers both the reliance trigger and the factors affecting the need to
protect the reliance of both the general contractor and the
subcontractor.
In sales contracts, particularly those involving complex transactions
(e.g., corporate acquisitions), information about the asset will be imper-
fect. Often, the most efficient producer of some types of information will
be the seller. The seller wants the buyer to rely on the information. The
greater the assurance, the more a buyer will be willing to pay. The seller,
therefore, has an incentive to provide information, perhaps in the form
of a warranty. However, the seller also wants to distinguish between infor-
mation that buyers should rely on, and, of equal importance, infor-
mation that buyers should not rely on. Part V explores issues concerning
reliance and information: antireliance clauses, breach of warranty, and
sandbagging.
I. RELIANCE AND FLEXIBILITY
Contract performance takes place over time, and the nature of the
parties future obligations can be deferred to take into account changing
circumstances. Reliance matters in this context since one or both of the
parties might want to rely on the continuity of the arrangement while, at
the same time, maintaining the flexibility to adapt as new information
becomes available. If the two parties are under single ownership, the
owner can determine the appropriate tradeoff between reliance and
flexibility. If they are not, the coordination is done by contract, and the
contract will define the tradeoff. The contract, as is often the case, might
give one party the discretion to adapt to the new information, and it
would convey to that party the counterpartys reliance, the cost to it of
granting that discretion. In effect, one party sells discretion (or flexibil-
Stick? The Economics of Promissory Estoppel in Preliminary Negotiations, 105 Yale L.J.
1249, 1250 (1996) (analyzing promissory estoppel doctrine as economic-regulation-
shaping bargaining process); Charles L. Knapp, Reliance in the Revised Restatement: The
Proliferation of Promissory Estoppel, 81 Colum. L. Rev. 52, 5254 (1981) (discussing
expansion of promissory estoppel in Restatement (Second) of Contracts and predicting
future proliferation of doctrine); Juliet P. Kostritsky, The Rise and Fall of Promissory
Estoppel or Is Promissory Estoppel Really as Unsuccessful as Scholars Say It Is: A New Look
at the Data, 37 Wake Forest L. Rev. 531, 53743 (2002) (surveying promissory estoppel
case law and criticizing as premature announcement of doctrines demise).
17. 333 P.2d 757 (Cal. 1958) (en banc).
18. See infra notes 207229 and accompanying text (discussing courts interpretation
of reliance under Drennan framework).
ity) to the other. The price would reflect the value of flexibility to one
party and the cost of providing that flexibility to the other. The price
need not be explicit. As some of the illustrations below will demonstrate,
the contract structure will determine an implicit price of flexibility.
The reliance-flexibility tradeoff could take the form of allowing one
party to terminate the agreementessentially giving it an option to
abandon. Breach is, of course, a special case of the option to terminate.
Once this is recognized, it is clear that the price of that option need bear
no relationship to the traditional remedies for breach. Or the tradeoff
could take the lesser form of, say, giving one party control of the quantity
decision. Both of these manifestations of the tradeoff will be explored in
Part I.
A. The Option to Abandon

Oliver Wendell Holmes famously said that the duty to keep a con-
tract at common law means a prediction that you must pay damages if
you do not keep it,and nothing else.19 The notion that a contract is
merely an option to perform or pay damages has elicited much criti-
cism. 20 A common refrain in both decisions and scholarship is that
parties plan for performance, not the option to terminate.21 That may be
true for the parties, but it is not true for their counsel. For many transac-
tions it would amount to malpractice to ignore the possibility that one or
the other party might choose to walk away from the deal. In contracts
where performance is expected to take place over time, there is a trade-
off between reliance and flexibility. On the one hand, the parties want to
adapt as new information becomes available. But on the other hand, if
one party has relied on the continuation of the relationship, the
counterpartys adaptation to that new information might cause it consid-
erable harm.
When circumstances change, sometimes the most efficacious
response is to terminate the contract. One party might have an option to
terminate while the counterparty might want to restrict that option to
protect its reliance by requiring payment. The exercise price for the
option to terminate need not have any relationship to the legal remedies
of the U.C.C. or Restatement (Second). In the contracts literature, the
notion of efficient breach has been controversial.22 The terminology is
19. O.W. Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 462 (1897).
20. See, e.g., Fried, supra note 9, at 1417 (arguing contract is promise that promisor
has moral duty to perform).
21. See 3 Samuel Williston, A Treatise on the Law of Contracts 1357 (1st ed. 1920)
(Parties generally have their minds addressed to the performance of contractsnot to
their breach or the consequences which will follow a breach.).
22. See, e.g., Melvin A. Eisenberg, Actual and Virtual Specific Performance, the
Theory of Efficient Breach, and the Indifference Principle in Contract Law, 93 Calif. L.
Rev. 975, 9971016 (2005) (arguing theory of efficient breach results in inefficiencies and
cannot be sustained); Daniel Friedmann, The Efficient Breach Fallacy, 18 J. Legal Stud. 1,
unfortunate. When framed as a design problem, it becomes a matter of

pricing the option to terminate. The manner in which reliance affects
the exercise price of the termination option varies depending on the
context.
To illustrate the variety, this section will describe a number of differ-
ent instances in which the price of the termination option is set. The
illustrations include the venture capital contract, the illusory contract,
the preliminary agreement, the breakup fee in a corporate acquisition,
and the pay-or-play contract in the movie industry. Despite the fact that
the reliance interest in these examples is substantial, the exercise price
can be quite low. Indeed, in some instances, the price is zero. The variety
of ways of pricing the termination option has implications for contract
remedies. This section concludes by considering such implications.
1. Venture Capital. Consider first the venture capital contract.23
The venture capitalist (VC) provides funds to an entrepreneur whose
project typically has a high risk of failure and, even if successful, a long
period before it will yield positive returns. The VC does not commit to
funding the entire project. Instead, it will phase payment, allowing it to
terminate its obligation as new information on the likely success of the
project becomes available. It pays for this option up front in the share
price. The option to abandon is generally accompanied by a right of first
refusal. If the entrepreneur finds an alternative supplier of funds, the VC
would have the right to continue funding on the same terms proposed by
the third party. The option to terminateto not throw good money after
badis valuable to the VC. But it is costly to the entrepreneur who would
be sitting with an incomplete project and no viable funding source to
complete it. The entrepreneur would be especially vulnerable were the
VC to use the option to rewrite the deal on more favorable terms. That is,
when the option to abandon is triggered, the VC could say that it would
only fund the next round if the terms were altered in its favor. So, what
protection of the entrepreneurs reliance would the contract exhibit? In
virtually all venture capital deals the answer is that the VC would pay
nothing at the time the option is exercised. All the entrepreneur has is
an unenforceable understanding that if the business performs as
expected, the VC will invest in another round. The entrepreneurs
protection of its reliance would take two forms. First is the VCs self-inter-
est: If the information produced has been positive, it will be in the VCs
2 (1989) (arguing efficient breach generates large expenses rather than reduces
transaction costs); Ian R. Macneil, Efficient Breach of Contract: Circles in the Sky, 68 Va.
L. Rev. 947, 94950 (1982) (noting fallacies underlying efficient-breach analysis); Joseph
M. Perillo, Misreading Oliver Wendell Holmes on Efficient Breach and Tortious
Interference, 68 Fordham L. Rev. 1085, 1090 (2000) (exploring misunderstandings of
efficient breach after Holmes).
23. For a discussion of the economics of venture-capital contracting, see generally
Ronald J. Gilson, Engineering a Venture Capital Market: Lessons from the American
Experience, 55 Stan. L. Rev. 1067, 107692 (2003).
interest to proceed. Delay or abandonment hurts it. Second, if the VCs

threat to terminate is perceived as an opportunistic attempt to renegoti-
ate, its reputation will suffer. For present purposes, the key point is that
although the entrepreneur relies on the VCs future funding, it would
rarely, if ever, receive compensation if the VC were to exercise the termi-
nation option.
2. Reliance on Illusory Promises. In some instances, the reliance-
flexibility tradeoff can be accomplished despite the parties deliberately
making their agreement legally unenforceable. That is, one party might
be encouraged to make investments in reliance on the continuation of its
relationship notwithstanding the fact that it would have no legal recourse
were the other party to terminate. If a contract said in effect, I will do it
if I want to, it would be deemed illusory and lacking consideration. For
example, the standard franchise agreement between Ford and its dealer-
ships pre-1940 took this form. The unenforceability was not an accident
of careless drafting. Ford knew what it was doing.24 Ford had only prom-
ised to fill future orders if it chose to do so. In Bushwick-Decatur Motors,
Inc. v. Ford Motor Co., the district court described a Ford franchise agree-
ment in detail:
The Sales Agreement, by its term Article (9)(c) was terminable
at any time, at the will of either party, and does not bind
defendant to sell or deliver, or plaintiff to buy, nor does it
impose any liability on defendant if it terminates or refuses to
make sales to the other party. The Sales Agreement was to be
observed by the parties only so long as was mutually
satisfactory.25
There, the court found the agreement to be illusory, despite the claimed
reliance of the dealer on alleged oral statements by Ford representatives:
[T]he oral contract in substance provided . . . that defend-
ants settled policy was Once a Ford dealer, always a Ford
dealer; that by the dealership contract the plaintiff had
become a member of the great Ford family; that the plaintiff
would remain a Ford dealer as long as it wanted to; that the
Ford policy, settled for many years, was a guarantee of this; and
that the plaintiff need have no hesitation whatever in investing
all available funds in the promotion of the sale and servicing of
Ford products as such investments would be perfectly safe. . . .
[T]he agreement was reaffirmed [numerous times and] at all
such occasions, plaintiff was encouraged to enlarge its facilities,
increase its sales force and expand its business, in reliance on
the assurances given by the defendant that plaintiff was in as a
Ford dealer as long as it wanted and should have no concern
24. See Friedrich Kessler, Automobile Dealer Franchises: Vertical Integration by

Contract, 66 Yale L.J. 1135, 115052 (1957) (explaining automobile companies
intentionally made franchise contracts unenforceable to thwart dealer lawsuits).
25. 30 F. Supp. 917, 92021 (E.D.N.Y.), affd, 116 F.2d 675 (2d Cir. 1940).
over the wisdom of making long term commitments and long

term plans.26
Neither Chrysler nor General Motors (GM) went to the extreme of mak-
ing their agreement unenforceable. However, by making the agreement
terminable on very short (say, fifteen days) notice, they essentially acc-
omplished the same thing.27 Their agreements were enforceable, but
only for the brief period. The auto manufacturers wanted their dealers to
make investments in reliance on continuation of the relationship, and by
and large dealers did so. The dealers relied on the expectation that their
satisfactory performance would assure renewal of their franchise. Dealers
wanted more but, absent legislative intervention, they could not get the
producers to give explicit protection for their reliance.28
The GM-Fisher Body contract, subject of much academic interest,29
provides another illustration. In 1919, the two entered into a ten-year
agreement in which Fisher agreed to produce and GM agreed to
purchase almost all of GMs closed automobile bodies.30 Despite the
considerable reliance by both, the agreement was unenforceable.31 While
GM promised to place orders for substantially all its bodies, Fisher only
promised to tell GM whether it would accept the orders. It did not
promise that it would fulfill them.32 The reliance of each was protected in
part by the consequences if either walked away. Both would have suffered
significant losses since neither had an adequate alternativethe high
switching costs, arguably, constrained the parties and protected the reli-
ance of each.33
Kelloggs contract with a supplier of packaging material for its cereal
provides a more current illustration of a clearly unenforceable agree-
ment.34 The contract was for three years.35 The quantity clause, such as
26. 116 F.2d at 678 (quoting Bill of Particulars).

27. See Ellis v. Dodge Bros., 246 F. 764, 765 (5th Cir. 1917) (indicating agreement
terminable with fifteen days notice); Buick Motor Co. v. Thompson, 75 S.E. 354, 355 (Ga.
1912) (describing agreement terminable with ten days notice).
28. Since passage of the Dealers Day in Court Act of 1956, automobile franchise
agreements that were not enforceable or were terminable on short notice have been
prohibited. Stewart Macaulay, Law and the Balance of Power: The Automobile
Manufacturers and Their Dealers 96103 (1966).
29. For a partial listing of the literature, see generally Victor P. Goldberg, Lawyers
Asleep at the Wheel? The GM-Fisher Body Contract, 17 Indus. & Corp. Change 1071
(2008) [hereinafter Goldberg, Asleep].
30. Id. at 107273.
31. See id. at 1076 (discussing unenforceability of contract).
32. See id. at 1075 (In the event that the FISHER COMPANY accepts the orders
specified in the schedules from time to time furnished by GENERAL MOTORS, it shall
proceed to make and deliver the automobile bodies called for by said schedules . . . .).
33. There is some question as to whether the executives of the firms were aware of the
unenforceability of their agreement. It is not clear that the arrangement was sustainable;
the two firms were merged before the ten-year period expired. Id. at 1072.
34. Complaint Exhibit 1, Kellogg Co. v. FPC Flexible Packaging Corp., No. 1:11-cv-272
(W.D. Mich. Mar. 18, 2011).
it was, read as follows: Kellogg generally encourages its employees to

obtain required goods and services from suppliers who have entered into
formal agreements with Kellogg, and Kellogg agrees to use reasonable
efforts to communicate the existence of this Agreement to such employees
with a general need to obtain the Products and Services within the scope
of this Agreement.36 Either party could terminate the Agreement with
120 days notice.37 While the agreement was, obviously, too open-ended
to be enforceable for any future orders, it would have been enforceable
for any orders that had already been executed. It, like the Fisher Body
contract and the Ford franchise contracts, was backward-enforceable, but
not forward-enforceable. Nonetheless, it likely provided sufficient assur-
ance to encourage the suppliers reliance on continuation of its dealings
with Kellogg.
3. Preliminary Agreements. Parties often enter into preliminary
agreementsletters of intent (LOIs), memoranda of understanding
(MOUs), and so forthprior to entering into a fully enforceable con-
tract. With such agreements, parties need not fully commit to a project.
They can defer their decision by waiting until further information is
revealed about the project and about their prospective partner. Rather
than negotiating an entire agreement, they can temporize with one or
both parties maintaining an option to terminate. The price of that
option will reflect the other partys reliance. The traditional common law
rule left reliance on preliminary agreements unprotected.38 In recent
years, following Judge Levals decision in Teachers Insurance & Annuity
Assn of America v. Tribune Co.,39 reliance has been treated as both a
rationale for enforcement and a remedy. Leval stated, Giving legal
recognition to preliminary binding commitments serves a valuable func-
tion in the marketplace, particularly for relatively standardized
transactions like loans. It permits borrowers and lenders to make plans in
reliance upon their preliminary agreements and present market
conditions.40
35. Id. at 3.
36. Id. (emphasis added).
37. Id.
38. See E. Allan Farnsworth, Precontractual Liability and Preliminary Agreements:
Fair Dealing and Failed Negotiations, 87 Colum. L. Rev. 217, 26369 (1987) [hereinafter
Farnsworth, Precontractual Liability] (Courts have traditionally accorded parties the
freedom to negotiate without risk of precontractual liability.); Friedrich Kessler & Edith
Fine, Culpa in Contrahendo, Bargaining in Good Faith, and Freedom of Contract: A
Comparative Study, 77 Harv. L. Rev. 401, 41220 (1964) (noting case law leaves parties
free to break off preliminary negotiations without being held to an accounting). See
generally Alan Schwartz & Robert E. Scott, Precontractual Liability and Preliminary
Agreements, 120 Harv. L. Rev. 661 (2007) (describing cases in which courts enforce
preliminary agreements and offering model for why parties make preliminary
agreements).
39. 670 F. Supp. 491 (S.D.N.Y. 1987).
40. Id. at 499.
He suggested that there were two types of enforceable preliminary

agreements. In Type I agreements, all the terms had been settled and
only the formality of a signed agreement was lacking.41 These agreements
would be enforced as contracts. His innovation was to recognize a lesser
commitment. In Type II agreements, a number of terms remained open
so the agreement could not be enforced as such.42 However, because the
parties had relied upon the negotiations, he wrote, there was a good faith
obligation to attempt to negotiate the deal to completion. If a party
breached the good faith obligation, it would be liable for damages.
Tribune was one of a trilogy involving the Teachers Insurance and
Annuity Association (TIAA).43 In the other two cases, the remedy was
expectation damages.44 The Tribune trial was only on the liability issue,
with damages to be determined at a subsequent proceeding.45 The case
settled, but it is likely that the settlement reflected the expectation
damagesessentially the difference in the present value of the loan at
the contract rate and the market rate. In subsequent Type II cases, most
courts have restricted recovery to reliance damages, following the reason-
ing of Allan Farnsworth: An award based on [the expectation interest]
would give the injured party the benefit of the bargain that was not
reached. But if no agreement was reached and it cannot even be known
what agreement would have been reached, there is no way to measure
lost expectation.46
If the preliminary agreement were silent on enforceability, the court
would have to determine, as a matter of law, whether a good faith duty
existed. Then the trier of fact would have to determine whether the party
had acted in good faith and, if not, what the damages would be. Of
course, the agreement could make the good faith requirement explicit,
and it could also determine the consequences of termination.47
41. Id. at 498.

42. Id.
43. Teachers Ins. & Annuity Assn of Am. v. Ormesa Geothermal, 791 F. Supp. 401
(S.D.N.Y. 1991); Teachers Ins. & Annuity Assn of Am. v. Butler, 626 F. Supp. 1229
(S.D.N.Y. 1986).
44. Ormesa Geothermal, 791 F. Supp. at 415; Butler, 626 F. Supp. at 1236.
45. Tribune, 670 F. Supp. at 508 (awarding judgment on merits without discussing
damages).
46. 1 E. Allen Farnsworth, Contracts 3.26a, at 386 (3d ed. 2004) [hereinafter
Farnsworth, Contracts]. Professor Eisenberg argues that the remedy should be the full
expectation interest, unless that is too uncertain, in which case the court should revert to
reliance damages. Melvin Aron Eisenberg, The Emergence of Dynamic Contract Law, 88
Calif. L. Rev. 1743, 1809 (2000).
47. In SIGA Technologies., Inc. v. PharmAthene, Inc., the parties agreed to negotiate in
good faith with the intention of executing a definitive License Agreement in accordance
with the terms set forth in the License Agreement Term Sheet. 67 A.3d 330, 33738 (Del.
2013). After the value of the license increased dramatically, the licensor regretted the
terms and proposed much different terms in the subsequent negotiationsmore than
doubling the royalty rate, for example. Id. at 33940. The court held that its negotiating
behavior was not in good faith and that, had it behaved in good faith, the parties would
In the leading New York case finding a Type II agreement, Brown v.

Cara, the agreement was silent on termination,48 although the subse-
quent negotiations suggested that the parties could easily have priced the
termination option. A construction company (Brown) and property
owner (Cara) entered into an MOU in which the construction company
would engage in an effort to have the property rezoned.49 It would also
get the construction contract and have partial ownership of the project,
but these terms were not spelled out.50 After success in rezoning, the par-
ties could not come to an agreement, and the property owner walked
away.51 The court held that this was a Type II agreement and that, if the
owners decision was not in good faith, then Cara would have to pay
damagespresumably reliance damages (Browns costs). 52 Since the
negotiations lasted for over a year and included multiple drafts of term
sheets and agreements,53 it is likely that a factfinder would have con-
cluded that Cara had negotiated in good faith.
Brown spent considerable resources in pursuit of the rezoning, rely-
ing on its expectation that it would do the construction and share in the
ownership of the completed project.54 Cara had the option to terminate,
but the MOU did not set an option price. The decision effectively priced
the option at Browns reliance damagesexpenses incurred (adjusted
for the probability that a court might find the termination to be in good
faith). But there were plausible alternative ways of pricing Browns reli-
ance. The MOU could have included a nonbinding clause, effectively
pricing the option at zero.55 The option price could have been based on
some fraction of the out-of-pocket reliance costs. Or the option price
could have made Browns compensation contingent on the success of the
deal.
One device for pricing a partys reliance is a buy-sell (or put-call)
arrangement. If, after some defined milestone (perhaps following the
have agreed to the terms in the Term Sheet. Id. at 347. It then awarded expectation
damagesthe projected stream of payments based on the terms in the Term Sheet. Id. at
35152.
48. 420 F.3d 148, 155 (2d Cir. 2005). For more detail on the case, see generally Victor
P. Goldberg, Brown v. Cara, the Type II Preliminary Agreement, and the Option to
Unbundle, in Conference on Contractual Innovation: Papers Presented (May 3, 2012)
[hereinafter Goldberg, Option to Unbundle] (unpublished manuscript) (separately
paginated work), available at http://web.law.columbia.edu/sites/default/files/microsites
/contract-economic-organization/files/Contractual%20Innovation%20book.pdf (on file
with the Columbia Law Review).
49. Brown v. Cara, 420 F. 3d at 151.
50. Id.
51. Id. at 152.
52. Id. at 15459.
53. Id. at 152.
54. Id. at 158.
55. Alternatively, the MOU could have stated that the parties would bear all of their
own precontract expenses.
rezoning), either party wanted to go it alone, it could trigger a buy-sell

that would give its counterpart a choice. The offeror would name a price
at which it would be willing to either take full ownership of the property
or sell out, and the offeree would choose. That would give both parties a
piece of the upside, but would still leave them with the flexibility to adapt
as they learned information about the project and about each other. In
fact, in the subsequent negotiations in Brown v. Cara, drafts of documents
included both a nonbinding clause and a buy-sell clause:
Either member (the initiating Member) may force the other
Member to either purchase the Initiating Members interest or
sell its interest to the Initiating Member at a price proposed by
the Initiating Member at any time if there is a deadlock on an
issue as to which the consent of both Members is required or
after the 3d anniversary of completion of Construction.56
The case illustrates how the option to abandon in the negotiating phase
can be valuable to one or both parties, that there are a wide variety of
plausible means for pricing that option (including zero), and that the
parties are perfectly capable of figuring out how to do so. The price
would reflect the ex ante reliance concerns of the parties.
4. Breakup Fees. In a corporate acquisition, there can be a tem-
poral gap of many months between execution of the contract and its
actual closing. During that period a lot can happen that might result in
the buyer wanting to terminate the deal.57 The terms governing the
buyers option to terminate are the subject of extensive bargaining. The
buyer is particularly concerned about the accuracy of the sellers repre-
sentations (adverse selection) and postexecution behavior (moral
hazard). It also would like the ability to walk away from the deal if
external forces result in a decline in the value of the seller or if it is
unable to arrange financing.
However, once the contract has been executed, the seller can have a
substantial reliance interest in the deal closing. The sellers reliance has
two components. First, there is the simple reliance on the agreed price. It
does not want to give the buyer the option to buy at the contract price
only if the market stays the same or goes up. Second, once the contract is
executed, the standalone value of the sellers business might be adversely
affected. Key personnel might start looking for new jobs; investment deci-
sions might be postponed; relations with old customers, suppliers, and
bankers might be impaired; and a failure to close might send a negative
signal about the viability of the company. The greater the sellers reliance
on the deal going through, the more protection (the higher the effective
option price) it will try to get.
The acquisition agreement requires that all the closing conditions
be met; if not, the buyer would have a zero-price option to walk away. Not
56. Goldberg, Option to Unbundle, supra note 48, at 56.

57. Sellers also want termination rights, but I will ignore those.
every inaccuracy, however trivial, would allow the buyer to walk away; typ-
ically, the walk-right is conditioned on the inaccuracies that individually,
or in the aggregate, constitute a material adverse effect (MAE).58 In addi-
tion, the closing can be contingent on a broader condition: Between the
exercise and closing date, there cannot have been a material adverse
change (MAC).59 The greater the sellers reliance, the more likely it is
that the MAC will be hedged by exceptions making it more difficult for
the buyer to walk.60 The exceptions generally put the risk of exogenous
change on the buyer, with the seller bearing the risk of endogenous
change (its behavior reducing the value of the combined firm) and its
inaccurate statements.
Some buyers, particularly private equity firms, include explicit
breakup fees in their agreements.61 If completely unconstrained, these
would put the risk of a decline in value from exogenous causes entirely
on the seller. Or the breakup fee could be conditional. The agreement
58. See Mergers & Acquisitions Mkt. Trends Subcomm., Am. Bar Assn, 2011 Private
Target Mergers & Acquisitions Deal Points Study 6065 (2012) (discussing frequency of
materiality qualifications on sellers representations and warranties in 2010 transactions).
59. In Delaware, the hurdle for proving a MAC is extremely high. See Frontier Oil
Corp. v. Holly Corp., No. Civ.A. 20502, 2005 WL 1039027, at *32*37 (Del. Ch. Apr. 29,
2005) (employing preponderance-of-evidence standard for proving material adverse
effect). The dearth of reported cases is somewhat misleading. If the existence of a MAC is
clear, the case will most likely not be litigated. Diamond Foods aborted acquisition of
Pringles provides one illustration of a deal falling through because of a MAC. Steven M.
Davidoff, Diamond Foods Debacle May Crack Open a MAC, N.Y. Times: Dealbook (Feb. 9,
2012, 11:36 AM), http://dealbook.nytimes.com/2012/02/09/diamond-foods-debacle-may-
crack-open-a-mac/ (on file with the Columbia Law Review).
60. In their study, Gilson and Schwartz considered the following exceptions:
(1) changes in global economic conditions; (2) changes in U.S. economic
conditions; (3) changes in global stock, capital, or financial market conditions;
(4) changes in U.S. stock, capital, or financial market conditions; (5) changes in
the economic conditions of other regions; (6) changes in the target companys
industry; (7) changes in applicable laws or regulations; (8) changes in the target
companys stock price; (9) loss of customers, suppliers, or employees; (10)
changes due to the agreement or the transaction itself; and (11) a miscellaneous
category. The agreements also were coded for two qualifications to the explicit
inclusions or exclusions of the traditional MAC definition. These qualifications
would make a specified inclusion or exclusion inapplicable if the MAC either
specifically affected the target company or had a materially disproportionate
effect on the target company.
Ronald J. Gilson & Alan Schwartz, Understanding MACs: Moral Hazard in Acquisitions, 21
J.L. Econ. & Org. 330, 34950 (2005).
61. E.g., United Rentals, Inc. v. RAM Holdings, Inc., 937 A.2d 810, 82026 (2007)
(detailing reverse breakup fee negotiations between private equity firm and acquisition
target in context of failed merger); see also Gerard Pecht & Mark Oakes, M&A Reverse
Breakup Litigation, Securities Law360 (Feb. 25, 2008, 12:00 AM), http://www.law360.com/
articles/48141/m-a-reverse-breakup-litigation (on file with the Columbia Law Review)
(Recently, many purchasers have successfully limited their exposure by capping their
liability at a pre-agreed reverse termination fee. Many recent deals, especially private
equity deals, expressly exclude specific performance as a remedy and provide that a
reverse breakup fee is the only remedy for a jilted target.).
in Hexion Specialty Chemicals, Inc. v. Huntsman Corp.62 provides a good

illustration. Because the value of the target firm had fallen, the buyer
wanted out.63 The buyers option had, in effect, three different prices. If
it could prove that there had been a MAC, it could walk away at a zero
price.64 If the buyer used its best efforts to obtain financing but failed, it
would pay a termination fee of $325 million.65 Finally, if the deal failed to
close because of a knowing and intentional breach of any covenant,
damages would be uncapped.66 Citing a number of bad acts by the buyer
and its lawyers, the vice chancellor found that the breach of a covenant
was intentional.67 After losing at trial, the buyer settled for $1 billion.68
The buyers right to terminate is not an afterthought; it is a signifi-
cant element in the deal and is often the subject of considerable
negotiation. The option priceor, as Hexion illustrates, the option
priceswould reflect the sellers reliance.
5. Pay-or-Play. Next, consider a movie studio contracting with an
actor to appear in a movie to be filmed in the near future (say, six
months later). That was the situation in the casebook favorite featuring
Shirley MacLaines dispute with Twentieth Century Fox following the
cancellation of the film project Bloomer Girl.69 A lot can happen in the
intervening period. The script might be disappointing; the genre might
become less attractive; a similar film might be released; a director
incompatible with the actor might sign on; a more attractive actor might
become available; the actors reputation might be tarnished in the
interim; and so forth. As the primary claimant on the projects earnings,
the studio has the incentive to make adaptive decisions that enhance the
expected value of the project. The right to terminate the actor before
filming commences would be valuable to the studio. The actor enters
into the contract in reliance on the project going forward and on her
being in the film. Her reliance is the opportunity cost of accepting this
project and forgoing other possible offers that might come along in the
intervening months. The studios flexibility and the actors reliance are
protected by a pay-or-play clause. The studio maintains the right not to
make the movie and, if it does choose to make the movie, to do so with-
out this actor. The price of this option depends on the marketability of
62. 965 A.2d 715 (Del. Ch. 2008).

63. See id. at 721 (After receiving the sellers first quarter 2008 results, the buyer . . .
began exploring options for extricating the seller from the transaction.).
64. Id. at 724.
65. Id.
66. Id.
67. Id. at 756.
68. Amy Kolz, The Big Fall: Wachtell, Lipton Was Supremely Confident of Its Strategy
in the Hexion Busted-Merger Litigation with Huntsman. So How Did It All Go So Wrong?,
Am. Law., Apr. 2009, at 68, 73.
69. Victor Goldberg, Framing Contract Law 279309 (2006) [hereinafter Goldberg,
Framing].
the actor. For a major talent, the pay-or-play option would kick in upon
entering into the contract and the compensation would be the so-called
fixed fee.70 For lesser talent, the option might not be triggered until a
subsequent eventperhaps the appearance of a bona fide alternative
offer for the actor. Until that point, the actor would be committed to the
project, but the studio would have a free option to terminate.
The pay-or-play clause is a variation on a severance package (except
the actor is not an employee). An employment agreement might give the
employer the discretion to terminate the agreement at its convenience
(no cause) and, if so, it might specify what compensation, if any, would
be required. The structure of these contracts can vary in subtle ways.
Consider, for example, the multiyear contracts of two coaches in major
college sports programs: John Calipari at Kentucky and Rich Rodriguez,
then at West Virginia. 71 The coach and the university both have a
substantial reliance interest in the relationship. If either chooses to
terminate without cause, that party must bear some consequence, but
both want to retain that option. The two contracts each choose a some-
what different way to protect reliance. If Kentucky were to terminate
Calipari, it would pay liquidated damages of $3 million per year for each
remaining year on the contract.72 Calipari would be required to make
reasonable and diligent efforts to obtain employment.73 If he were to
succeed, then the damages would be reduced by the amount of the
minimum guaranteed annual compensation package of the Coachs new
position.74 That is, he would have a duty to mitigate and there would be
a partial offset. If, on the other hand, Calipari were to choose to termi-
nate early, he would owe as liquidated damages an amount that would
decrease over time$3 million if the termination were in the first year,
70. Major talent typically receives a percentage of the gross offset against a fixed fee.
The fee might be in the $20 million range. If the share of the gross were 10%, the gross
would have to reach $200 million before the contingent compensation would kick in. For
more detail on contingent compensation in the movie business, see id. at 1342.
71. Employment Agreement Between the Univ. of Ky. and John Vincent Calipari
(Mar. 31, 2009) [hereinafter Calipari Contract] (on file with the Columbia Law Review);
Employment Agreement Between the Univ. of W. Va. and Richard Rodriguez (Dec. 21,
2002) [hereinafter Rodriguez Contract] (on file with the Columbia Law Review).
72. Calipari Contract, supra note 71, at 11. Caliparis base salary was only $400,000.
Id. at 4. However, his compensation also included a broadcast endorsement payment of
around $3 million per year. Id. at 5. The contract included incentive payments based on
the athletic and classroom performance of his team. Id. at 78. If Kentucky won the
Southeastern Conference and national championships, the incentive payment would be
an additional $750,000. Id. at 7. If the basketball team achieved a 75% graduation rate, he
would receive an additional $50,000. Id. at 8. Since most of his stars were one-and-done,
he seemed quite willing to sacrifice this piece of the compensation.
73. Id. at 12.
74. Id. If the new contract were structured in the same way, it appears that this would
only refer to the base compensationabout $400,000, even though the contract would
likely pay out over $3 million per year.
declining to $500,000 in the fourth year and nothing in year five.75 The
Rodriguez arrangement was simpler. If either he or the school had termi-
nated without cause, there would have been a one-time payment of $2
million.76 Rodriguez would have had to neither mitigate nor offset any
earnings.77
The point of these examples is that the option to terminate the
agreement can create value, and the price of that option reflects the
counterpartys reliance. I say reflects, since, as some of the preceding
examples show, there can be considerable reliance but little or no com-
pensation if the option were to be exercised.
6. Remedies for Breach. If a contract does not explicitly allow for
termination, what then? The default rule is that termination would
amount to a breach and the promisor would be liable for damages (or
specific performance). The preceding discussion of how parties protect
their reliance from the possibility that the counterparty might terminate
the agreement has implications for contract remedies. Importantly, these
concern not only the reliance remedy, but the expectation remedy.
Specifically, I will argue that in many instances the benefit of the
bargain goes well beyond what sophisticated parties would (and in fact
do) choose.
Terminating an agreement is, of course, not the only way in which a
contract can be breached, and I will consider another in the next section.
The starting point is that the promisor has an option to terminate with
the option price being the remedy for breach. That framing has gener-
ated much criticismindeed, hostility. Breach is immoral to some and
the amorality of the option notion grates. Treating a contract breach as a
transgression against anothers rights, Friedmann would go beyond
expectation damages on grounds that a party is generally bound to per-
form his contractual promises unless he obtains a release from the
promisee.78 The remedy, he insists, should not be damages, but specific
performance.79 However, after pages of argument against allowing the
promisor to walk away, Friedmann takes almost all of it back:
As a normative matter, parties in a contractual setting should be
left free to define the ambit of their rights, and it is open to
75. Id. at 16.

76. Rodriguez Contract, supra note 71, at 10.
77. Id. In a typical pay-or-play contract, the talent has no duty to mitigate, but he or
she has to offset any earnings. That is explicit in the union contract of the Directors
Guild: If the director is employed by a third party, the employer shall be entitled to an
offset of the compensation arising from such new employment for such remaining portion
of the guaranteed period against the compensation remaining unpaid . . . . However, the
Director shall have no obligation to mitigate damages arising from his or her
removal . . . . Thomas D. Selz et al., 3 Entertainment Law 27.10, at 34 (2d ed. 2007)
(quoting Directors Guild of Am., Basic Agreement 6-105 (1993)).
78. Friedmann, supra note 22, at 2.
79. Id. at 7.
them to stipulate that the promisor will be allowed to terminate

the contract subject to the payment of damages. The efficient
breach theory assumes, however, that, even if they have not
done so and even if they intended to confer on the promisee a
broader entitlement, the law will nevertheless defeat their joint
intention by granting the promisor the option to breach.80
The parties should be free, that is, to choose what, if anything, should
happen if one party wants to terminate. Friedmanns entire attack is on a
default rule, and he appears quite content to allow parties to contract out
in any way they see fit. His argument boils down to the notion that spe-
cific performance should be the default remedy, not expectation dam-
ages. Ironically, Scott and Triantis, taking the options framework, reach
the same conclusion.81 Their rationale is quite different. Framing the
question in terms of options makes it clear that the option price need
bear no relationship to the standard damage remedies of contract law
expectation, reliance, or restitution.82 Rather than privileging any of the
three, Scott and Triantis recommend the none of the above
alternative.83
I regard the Scott and Triantis proposal as a form of academic shock
treatment. The expectation damages remedy is so firmly entrenched in
the minds of judges, lawyers, contracts professors, and even first-year law
students that a bold proposed change was necessary to get their atten-
tion. The fundamental point is that the contract law remedy is, in effect,
the implied termination clause, and that it should be viewed as just
another contract term from which parties are free to vary. The remedy
default rule, however, is stickier than others. Although most of contract
law provides default rules from which parties are free to contract away,
remedial defaults carry heavier presumptive weight than other provi-
sions.84 Doctrinal limitations, notably the unenforceability of penalty
clauses, present barriers to the proper pricing of the termination option.
The stickiness of the expectation damage remedy is, as Scott and
Triantis observe, in part a historical accident,85 but it also has great
80. Id. at 23.

81 . Scott & Triantis, Embedded Options, supra note 7, at 148688. Specific
performance is their default rule for contracts between sophisticated businesses. Id. For
consumer contracts, their default rule is to give the consumer a free option. Id. at 1488
90.
82. Id. at 145676.
83. Id. at 147786.
84. Id. at 1435. But there is little reason to accord greater weight to damage rules. See
Richard A. Epstein, Beyond Foreseeability: Consequential Damages in the Law of
Contract, 18 J. Legal Stud. 105, 108 (1989) (Damage rules are no different from any
other terms of a contract. They should be understood solely as default provisions subject
to variation by contract. The operative rules should be chosen by the parties for their own
purposes, not by the law for its purposes.).
85. Scott & Triantis, Embedded Options, supra note 7, at 1446 ([T]he preceding
inquiry into [the compensation principles] historical roots suggests that the elevation of
rhetorical power. If a breaching party is perceived as having wronged the

promisee, then corrective justice would seem to require that, like a tort
victim, the promisee should be made whole. The standard refrain is that
the contract remedy should put the promisee in as good a position
(financially) as if the contract had been performed. The promisee
should get the benefit of its bargain. That provides an anchor for doctri-
nal argument and friction for moving away from the default remedies.
Reframing the problem as a matter of transaction design and recogniz-
ing the reliance-flexibility tradeoff show why the benefit of the bargain
remedy is too simplistic. Holmess framing as the promisor having a
choice between performing or paying damages was, in large part, a
response to the notion that default rules should be derived from ethical
norms rather than commercial needs. Reviving Holmess aphorism
would at least nudge the rhetoric in a more useful direction. In particu-
lar, the penalty doctrine could be softened, if not abandoned entirely.
And, this Essay argues below, in some contexts awarding a seller lost
profits would overprotect its reliance.86
Much scholarship, particularly economic analyses of remedies,
proceeds on the implicit assumption that the parties are incapable of
pricing termination themselves and that the policy options are limited to
damages (expectation, reliance, restitution) or specific performance. By
framing the problem as one of pricing the termination option, it is clear
that in a wide variety of contexts the efficient rule (i.e., one that sophis-
ticated parties would voluntarily choose) bears no relation to the ones
featured in the lawyers rhetoric and the economists model. To be sure,
there are classes of cases in which a default remedy of expectation
damages would be efficient. I argue elsewhere that there is a strong case
for protecting the contract-market differential, which I have labeled the
narrow expectation interest.87 But beyond that, there is little reason to
believe that there would be any relationship between the option price
and the default remedies of law or the remedies modeled by economists.
To illustrate the point, consider one context in which the expecta-
tion remedy does a particularly poor job, the so-called lost-volume-
seller problem. The rule is memorialized in U.C.C. section 2-708(2),
which elevates economic misunderstanding into a statutory command:
compensation to a universally applicable norm results more from mistaken path

dependence than from a sustained and systematic appreciation of the merits of the rules
governing contract damages.).
86. See infra notes 88106 and accompanying text (providing examples of lost profits
creating overprotection).
87. Goldberg, Framing, supra note 69, at 21924. In a study of the use of reliance
damages in contracts under the U.C.C., Gibson found that courts rarely awarded reliance
damages. See Michael T. Gibson, Reliance Damages in the Law of Sales Under Article 2 of
the Uniform Commercial Code, 29 Ariz. St. L.J. 909, 915 (1997) (Thus, in 467 of the
cases most likely to produce reliance damages, only fourteen (2.9%) did so.). However, as
noted below, some parties do contract to limit the remedy to reliance damages. See infra
note 107 and accompanying text.
If the measure of damages provided in subsection (1) is inadequate to

put the seller in as good a position as performance would have done
then the measure of damages is the profit (including reasonable
overhead) which the seller would have made from full performance by
the buyer.88 In the lost-volume cases, the contract and market price are
the same, so if a buyer were to cancel an order, there would appear to be
no damages. The claim is that the seller should be compensated because
it would have had the additional sale and its lost profit would be the
difference between the contract price and its but-for costs. For a
retailer, that would be the gross marginthe retail minus wholesale
price. While the retailing cases play a prominent role in the literature
and pedagogy, with Neri v. Retail Marine Co.89 being the case of choice, in
practice, they are of little relevance. Since Neri was decided in 1972, there
have been only four recorded cases involving claims by retailers against
consumers. The retailer won in two cases (one for a car90 and one for a
boat91) and lost two (one for a boat92 and one for a mobile home93).
If the buyer walked away and the dealer sold the boat at the same
price, the rule focuses on whether the seller would have sold another
boat.94 If so, the seller has lost his profit (the wholesale-retail differen-
tial) on the second boat and would be entitled to a damage award of the
lost profits.95 The analysis gets more complicated when the sellers ability
to sell the second boat is questioned. If the seller would not have been
able to sell another unit, perhaps because the supply was limited (it
could not get another boat from the manufacturer, perhaps because this
model was so attractive), then there would be no damage.96 So goes the
rule. When framed as setting an option price, the remedy makes no
sense.
When placing an order, the customer wants some flexibility (the
option to change her mind) and the seller wants some protection of its
reliance on the order. The remedy prices the option. For cars and boats,
it would be in the 12% to 15% range. In effect, I agree to pay 15% of the
sale price for the option to buy the boat by paying the other 85%.
88. U.C.C. 2-708(2) (2012).

89. 285 N.E.2d 311 (N.Y. 1972).
90. Van Ness Motors, Inc. v. Vikram, 535 A.2d 510 (N.J. Super. Ct. App. Div. 1987).
91. Modern Marine, Inc. v. Balski, No. 43411, 1981 WL 4969 (Ohio Ct. App. May 21,
1981).
92. Lake Erie Boat Sales, Inc. v. Johnson, 463 N.E.2d 70 (Ohio Ct. App. 1983).
93. Schiavi Mobile Homes, Inc. v. Gironda, 463 A.2d 722 (Me. 1983).
94. U.C.C. 2-708(2) (2012); see also Lake Erie, 463 N.E.2d at 72 ([T]he controlling
issue regarding . . . entitlement to lost profits is whether the [dealer] sufficiently
established its status as a volume seller.).
95. See Neri v. Retail Marine Co., 285 N.E.2d 311, 314 (N.Y. 1972) (noting under
section 2-708, measure of damages for dealer with unlimited supply or standard-priced
goods is dealers profit on one sale).
96. Lake Erie, 463 N.E.2d at 72.
Leaving aside the question of whether any consumer would have any idea
that she had made such a commitment, we can ask whether an informed
consumer would be willing to pay such an option price. The answer
almost certainly is no. The U.C.C.s remedy overprotects the sellers reli-
ance, which is, typically, trivial. The dealer can take into account the
likelihood that a certain percentage of its orders will result in cancella-
tions. The likelihood that a buyer will walk away is one of the risks of
doing business. It is predictable; more importantly, it can be influenced
by the sellers decisions. In particular, the seller can set an explicit option
pricesay, a nonrefundable deposit. There is no reason to believe that
the option price would be equal to the lost-volume remedy. Indeed, the
remedy would be perverse. If the market were weak, the option price
would likely be near zero; conversely, if the market were tight, the option
price would be higher. The remedy gets it backwards. When the market
is tight (say the manufacturer allocates only a set number of boats to the
dealer), the seller would not be able to sell an extra unitthe lost-
volume remedy would be zero. When the market is slack, the U.C.C. rem-
edy would be the full wholesale-retail margin.97
In the B2B context, there is an even greater disconnect between the
remedy and its function. The aggrieved seller claims that, had there been
no breach, it would have been able to produce and sell all the other units
anyway; accordingly its loss would be the difference between the contract
price and the costs it would have incurred had it produced the units the
buyer refused to takethe but-for costs. Costs not associated with pro-
duction of these unitsresearch and development, advertising, market-
ing, most plant and equipmentwould not count; they would be part of
the lost profit. Lost profit, especially for high-tech products, could there-
fore be a significant part of the contract pricefor example, in Teradyne,
Inc. v. Teladyne Industries, Inc., the lost profits amounted to about 75% of
the contract price.98 The contract created by the lost-volume remedy
would have the buyer paying $75 for the option to purchase the product
for an additional $25. Such a deal makes little business sense, especially if
the seller has ample capacity to meet the needs of this buyer and others
(which must be so if the seller is indeed a lost-volume seller). As in the
retail context, the presumption would be that the option price would be
greater if the seller faced significant output constraintsthe remedy
again gets it backward.
97. For a fuller exposition of this argument, see Goldberg, Framing, supra note 69, at
23342; see also Scott & Triantis, Embedded Options, supra note 7, at 148284 (arguing
scholarly focus on lost volume and selling costs is a red herring).
98. See 676 F.2d 865, 867 (1st Cir. 1982) (In effect, this was a finding that Teradyne
had saved only $22,638 as a result of the breach. Subtracting that amount and also the
$984 quantity discount from the original contract price of $98,400, the master found that
the lost profit (including reasonable overhead) was $74,778.).
An interesting illustration is provided by Rodriguez v. Learjet, Inc.99

The contract for a new airplane included a series of progress payments;
in the event that the buyer terminated, Learjet would keep the payments
already made.100 The buyer terminated shortly after entering into the
agreement and sued for return of its initial payment of $250,000, assert-
ing that the payment was an unenforceable penalty clause.101 Learjet
defended by arguing that it was not a penalty.102 To show this, Learjet
argued that it was a lost-volume seller, demonstrating that it had the
capacity to sell this plane and another; it proved to the courts satisfac-
tion that it had suffered lost profits of over $1.8 million.103 Therefore, the
measly $250,000 was not a penalty and the court enforced it.104 The value
of getting approval of its progress payment schedule (a schedule of
option prices) apparently exceeded the one-time gain of $1.8 million.
The case illustrates one function of the lost-volume-profit remedy:
enforcing deposits. If the buyer were to sue for return of its prepayment,
the sellers counterclaim for lost profits would act as a deterrent.105 If
doctrine recognized that the prepayment was payment for the walk-
option, there would be no need for this indirect device for enforcing the
option price.
So, U.C.C. section 2-708(2) is perverse. What can we do about it?
One response is to interpret it narrowly; another is to enforce deposits
and other termination remedies. More generally, we should view the
analysis as a step in chipping away at the dominance of the notion that
contract damages should be thought of as compensation for a wrongful
breach rather than as a decision variable for the parties.
I do not expect contract law to abandon the benefit-of-the-bargain
approach. I do hope, however, that focusing on the contract-design
questions will have some effect. The lost-volume-seller problem is one
example. The notion that lost profits in general should routinely be
enforceable is another. The unenforceability of penalty clauses, discussed
below,106 is yet a third. Perhaps the analysis might nudge doctrine a bit in
the right direction by requiring a high standard of proof for the lost prof-
its on the one hand, and by facilitating the limitation of damages on the
other.
99. 946 P.2d 1010 (Kan. Ct. App. 1997).

100. Id. at 1012.
101. Id. at 1013.
102. Id.
103. Id. at 101415.
104. Id. at 1015.
105. Neri also was a counterclaim in response to the buyers suit for return of a down
payment. Neri v. Retail Marine Co., 285 N.E.2d 311, 312 (N.Y. 1972). However, these cases
appear to be an exception. Based on my own count, only around 10% of the reported
cases involve a counterclaim.
106. See infra note 135 and accompanying text (discussing unenforceable penalty
clauses).
B. Quantity Adjustment
When determining the quantity in a long-term contract for the sale
of goods, the parties confront a problem. Conditions on both the
demand and supply side can vary over time, and they might want to
adjust the quantity as new information becomes available. While Part I.A
discussed an adjustment to change involving termination, this section
concerns a more modest adjustmentquantity variation.
When the manufacturer, say, of automobiles enters into supply con-
tracts, it wants the flexibility to react as new information regarding the
demand for specific models becomes available. If the news is bad, the
manufacturer generally maintains the right to cancel part or all of the
order. For example, the standard General Motors Purchase Order Terms
and Conditions says:
Buyer may change the rate of scheduled shipments or direct
temporary suspension of scheduled shipments, neither of which
shall entitle Seller to a modification of the price for goods or
services covered by this order . . . .
. . . In addition to any other rights of Buyer to cancel or
terminate this order, Buyer may at its option immediately termi-
nate all or any part of this order, at any time and for any reason,
by giving written notice to Seller.107
If GM did exercise its option to cut back its order, the contract gave the
seller the equivalent of the standard reliance remedy. GM would pay for
all items that had already been completed under the purchase order and
the costs of work-in-progress less the value of any goods the supplier
could resell to third parties. There would be no recovery for overhead
and other elements of lost profits.108
107. Am. Axle & Mfg. Holdings, Inc., Amendment No. 1 to Form S-1 Registration
Statement Under the Securities Act of 1933 (Form S-1/A) (June 5, 1998) [hereinafter GM
Standard Form], available at http://www.sec.gov/Archives/edgar/data/1062231/0000889
812-98-001427.txt (on file with the Columbia Law Review) (including Purchase Order as
exhibit D to Component Supply Agreement Between American Axle & Manufacturing,
Inc. and General Motors Corp.). Other forms of the Original Equipment Manufacturers
(OEMs) have similar language. Major suppliers are referred to as Tier 1 suppliers, and
they, in turn, contract with Tier 2 suppliers. For example, Eberspaecher North America,
Inc. is a manufacturer of exhaust systems, which it sells to OEMs. Its standard Terms and
Conditions include similar language: In addition to any other rights of Buyer to
terminate the Order, Buyer may, at its option, terminate all or any part of the Order
without any liability whatsoever to Seller, at any time and for any reason, by giving 30 days
written notice to Seller. Complaint and Demand for Trial by Jury Exhibit 2 at 6,
Eberspaecher N. Am., Inc. v. Nelson Global Prods., LLC, No. 2:12-cv-11045-RHC-PJK (E.D.
Mich. Mar. 8, 2012).
108. GM Standard Form, supra note 107, cl. 13. The costs OEMs are willing to cover
vary:
While all OEMs provide some recovery to suppliers for their squandered
investments, some are stingythey pay only for finished parts, work in progress,
and raw materials. Others are more generous: they will pay for a combination of
other termination costs, such as suppliers obligations to their own
Long-term variable-quantity contracts are quite common. They are

even singled out in the U.C.C., which recognizes both full output and
requirements contracts.109 A full output contract gives the seller discre-
tionif I choose to produce any of this product, you promise to take all
of it. A requirements contract gives the discretion to the buyerI will
only take my requirements from you and you promise to be ready to sell
to me any amount that I desire. The counterparty would be reluctant to
make such an open-ended commitment, and the agreement would likely
include some limitation on that discretion. For example, the buyer might
agree to take all its requirements for the operation of a particular plant;
or it might agree to limit the sellers obligation to a monthly maximum.
The contract will give discretion to adapt to changed circumstances to
the party that values it most, and the price of that flexibility will reflect
the reliance of the counterparty. The contract could constrain the discre-
tion in a variety of ways. Unfortunately, the U.C.C. ignores any functional
limitation and instead uses the amorphous standard of good faith to
limit the discretion.
Consider a firm that, along with its primary product, produces a
byproduct that has some modest commercial value but if not removed
promptly will cause problems for the producer that might have to add
storage space or, in the extreme, curtail production of its primary prod-
uct. Examples include petroleum coke (petcoke), a byproduct of the
coking process for refining petroleum to produce gasoline and other
valuable products, and day-old bread, an inevitable byproduct of the
production and sale of fresh bread. The firm could enter into a long-
term contract to assure prompt removal of the byproduct. If the byprod-
uct had no economic value, this would simply be the equivalent of a
trash-removal contract; if, as in these examples, there is some economic
value, the contracting problem is more interesting.110
Petcokes commercial value is low, less than 3% of the value of the
oil.111 To produce petcoke, a refinery has to add a coker, a multimillion
dollar plant with no other use.112 The buyer would operate a calciner,
subcontractors and investments in capital. None of the OEMs cover suppliers

unamortized investment in R&D and engineeringa great source of agony for
suppliers who expect to cover their fixed costs only after several years of supply.
Omri Ben-Shahar & James J. White, Boilerplate and Economic Power in Auto
Manufacturing Contracts, 104 Mich. L. Rev. 953, 959 (2006) (footnotes omitted). Toyota
would compensate for the actual cost of Supplier for all works in process. Toyota Motor
Mfg. N. Am., Inc., Terms and Conditions art. 5.8(d), at 22 (Oct. 1, 1998) [hereinafter
Toyota T&C] (on file with the Columbia Law Review). Actual cost is not defined, but it
most likely excludes overhead costs and lost profit.
109. U.C.C. 2-306(1) (2012).
110. These illustrations, and others, are examined in more detail in Goldberg,
Framing, supra note 69, at 10144.
111. Id. at 105.
112. Id. at 104.
also a single-use, multimillion-dollar item.113 The primary use for the

calcined coke was to make anodes, which were used to produce alumi-
num.114 There were, to simplify things a bit, two different types of con-
tracts. In one, the refinery (seller) had very limited storage space and the
buyer had the capacity to take petcoke from a number of refineries and
hold it in inventory.115 The seller desired the freedom to make all its deci-
sions on the basis of the demand for gasoline and other refined prod-
ucts, not petcoke. Granting this flexibility was essentially costless to the
buyerthere was no reliance on this particular supplier. The resultant
contract was a full-output contract, which gave the seller complete
discretion as to how much petcoke it would produce; the buyer promised
to immediately remove the petcoke and bore all the risks of quantity
fluctuation.116
The second type of contract concerned the simultaneous construc-
tion of a coker and an adjacent calciner operated by an aluminum
producer.117 Because petcoke was expensive to transport, the buyers only
practical source would be the output of the new coker.118 To protect the
buyers reliance, the contract gave the quantity decision to the buyer.119
The contract specified a maximum amount, but allowed the buyer to
take less if it so chose.120 The buyers discretion was constrained to take
into account the refinerys reliance. For example, one contract included
a standby fee, which amounted to about 40% of the monthly quantity;
even if the buyer took no petcoke it would have to pay.121 Other contracts
with the aluminum producers had similar constraints on the sellers free-
dom to vary quantity.122
The standby fee is one variation on a take-or-pay contract. In a
take-or-pay contract, the buyer agrees to pay for a certain percentage of
the specified quantity, regardless of whether or not it actually takes it.
The price for the first, say, 20% of the product in any given month is, in
effect, zero. If the value of the sellers plant is contingent on the contin-
ued purchases by the buyer, the guaranteed take is one way to protect
the sellers reliance. The greater the reliance, other things equal, the
higher the guaranteed payment will be. The sellers reliance need not, in
general, be fully protectedthat is, the parties can share the risk of a bad
outcome by setting the sum of the guaranteed payments below the
sellers costs if the buyer were to order less than the minimum; however,
113. Id.
114. Id.
115. Id. at 105.
116. Id.
117. Id.
118. Id. at 106.
119. Id.
120. Id.
121. Id.
122. Id.
there can be scenarios in which the guaranteed payments would exceed

the cost. That possibility has made take-or-pay clauses (or a variation such
as a minimum quantity) subject to the penalty-clause doctrine. A signifi-
cant case in which a court (Judge Posner, to be precise) found a
minimum-quantity clause to be a penalty is Lake River Corp. v.
Carborundum Co.123 I will turn to that shortly, but first I want to discuss the
leading New York case regarding a variable-quantity clause, Feld v. Henry
S. Levy & Sons, Inc.124
Levy baked and sold rye bread and inevitably produced, as a byprod-
uct, stale and imperfect loaves.125 It decided to install an oven to convert
these into breadcrumbs, which had some commercial value.126 To assure
prompt removal, it entered into a one-year, full-output contract with
Feld.127 Either party could cancel the contract on six months notice.128
To protect its reliance, Levy required that Feld obtain a faithful perfor-
mance bond.129 Because Feld had other sources of breadcrumbs, the
contract put no constraints on Levys discretion.130 Things did not work
out as Levy had planned and it was losing money. It tried to renegotiate
the price, but Feld refused. Levy then dismantled the oven and ceased
producing breadcrumbs. Since the contract required that Levy give six
months notice for termination, the dismantling should not have been
viewed as a termination. Had Levy reinstalled the oven it would still have
been obligated to deliver all the crumbs to Feld.131 Feld sued for breach,
arguing that Levy could not in good faith reduce its output to zero.132
The court of appeals held that, despite the contract language, the
seller was not free to decide whether it should produce any breadcrumbs
at all.133 The implied duty of good faith required that it continue to pro-
duce breadcrumbs even if there be no profit. In circumstances such as
these and without more, defendant would be justified, in good faith, in
ceasing production of the single item prior to cancellation only if its
losses from continuance would be more than trivial, which, overall, is a
question of fact.134 The court failed to recognize that Levy was the one
making an investmentthe toaster ovenin reliance upon the contract.
The cost to Feld of granting the discretion was zero. The court, in effect,
123. 769 F.2d 1284, 128687 (7th Cir. 1985).

124. 335 N.E.2d 320 (N.Y. 1975). For more detail on this case, see Goldberg, Framing,
supra note 69, at 11719.
125. Feld, 335 N.E.2d at 321.
126. Goldberg, Framing, supra note 69, at 117.
127. Feld, 335 N.E.2d at 321.
128. Id.
130. Feld, 335 N.E.2d at 321.
132. Id.
133. Id. at 118.
134. Feld, 335 N.E.2d at 323.
held that Levy promised that it would be willing to lose some money
but not too muchin order to protect Felds nonexistent reliance. The
contract priced Levys discretion at zerothe court trumped it for no
good reason.
In Lake River, Judge Posner ruled that a minimum-quantity clause
was an unenforceable penalty.135 Lake River agreed to bag an abrasive
powder produced by Carborundum.136 They entered into a three-year
agreement with a minimum of 22,500 tons over the life of the contract.137
Lake River installed bagging equipment costing $89,000 to be used exclu-
sively for Carborundum.138 The decision did not note one feature of the
contractLake River promised that it would bag up to 400 tons per
week; Carborundum could conceivably have asked Lake River to bag over
60,000 tons during the life of the contract.139 The contract was not exclu-
sive on either sideboth were free to deal with others during the
contract period. 140 Times were tough for Carborundum, and it only
managed to deliver about 12,000 tons.141 Lake River sued, claiming that it
should be paid for the 10,500 tons that Carborundum had failed to
deliver.142 The case was framed in terms of whether the suspect clause
was for liquidated damages or a penalty.143 Carborundums defense was
that it was an unenforceable penalty clause, and Judge Posner agreed.144
He noted that any shortfall would have left Lake River with a greater
profit than if the minimum had been reached.145 If the breach had
occurred on the first day, he noted, the damages would have been over
five times the investment that Lake River was making in special bagging
equipment.146
Instead of framing the matter in terms of liquidated damages versus
penalty, it is more productive to ask why the agreement was structured in
that way. Carborundum was given substantial discretion as to whether it
should use Lake Rivers facility and, if so, how much. The contract effec-
tively set a fixed fee and a price for the first 22,500 tons of $0. Lake River
promised that during the three-year period, it would make available
135. Lake River Corp. v. Carborundum Co., 769 F.2d 1284, 1290 (7th Cir. 1985); see
generally Victor P. Goldberg, Cleaning Up Lake River, 3 Va. L. & Bus. Rev. 427 (2008)
[hereinafter Goldberg, Lake River] (providing facts about Lake River not included in
published opinion).
136. Lake River, 769 F.2d at 1286.
137. Id.
138. Id.
139. Goldberg, Lake River, supra note 135, at 439.
140. Id. at 440.
141. Lake River, 769 F.2d at 1286.
142. Id.
143. Id. at 1288.
144. Id. at 1290.
145. Id.
146. Id. at 1292.
capacity to bag 400 tons per week at the contract price.147 The court
treated Lake Rivers investment in the new bagging equipment as deter-
mining the outer boundary of its reliance. However, Lake River also had
to have workers on hand to handle any product delivered (up to the
weekly maximum); in addition, it faced the opportunity cost of holding
the capacity ready for the entire period. The minimum quantity indi-
rectly set a price for the flexibility. Did the deal overprice
Carborundums discretion? Ex post, yes. But ex ante, it is not so clear.
Perhaps Carborundum did pay too much for the flexibility, but there is
no reason to second-guess the consideration paid for this valuable
service. At the end of the three years, Lake River had fully performedit
had remained ready, willing, and able to take 400 tons per week for the
entire period. All that remained to be performed was the payment by
Carborundum.
The fundamental point in all these examples is that discretion can
be allocated to the party that values it most and the counterparty can pro-
tect its reliance by, in effect, charging a price for the flexibility. By failing
to recognize how parties resolve the reliance-flexibility tradeoff, contract
doctrine can get in the way, using concepts like good faith (Feld) and
penalty clauses (Lake River) to thwart the parties intentions. The greater
the reliance, other things equal, the higher the price will be. That the
price is not quoted explicitly matters not.148
II. RELIANCE AND CONSEQUENTIAL DAMAGES
The role of reliance in a consequential damage claim is different. In

a typical case, one party, A, relies on the successful performance by the
other, B, but is disappointed. Perhaps B was negligent, as in Hadley v.
Baxendale.149 Or perhaps B was not at fault, but for reasons beyond its
control, B failed to perform.150 In either instance, A might have a claim
for compensation for losses it suffered as a consequence of its reliance.
147. Goldberg, Lake River, supra note 135, at 439.

148. These examples have all concerned reductions in quantity, but similar problems
can arise when the seller wants to increase quantity. The contract can have a maximum
(like Lake Rivers 400-ton weekly maximum); if the seller wants to provide more, the
contract could bar it, which would give the buyer a baseline from which to negotiate a
price for the increased quantity. There are a lot of alternatives. For example, in a contract
between Columbia Nitrogen and Royster (a casebook staple), the buyer could ask for
more if the seller had the capacity to provide it. Columbia Nitrogen Corp. v. Royster Co.,
451 F.2d 3, 6 n.2 (4th Cir. 1971). But it could only ask. The contract did not prevent the
seller from using the capacity to sell to others, so the ability to sell to others, in effect,
meant that the additional quantity could be sold at the current market price, rather than
at the contract price. Goldberg, Framing, supra note 69, at 16465.
149. (1854) 156 Eng. Rep. 145 (Ex.); 9 Ex. 341.
150. The recent English case that reconsidered the Hadley doctrine, The Achilleas,
concerned losses arising from late redelivery of a legitimate last charter, the delay being
the fault of neither party. See generally Victor Goldberg, The Achilleas: Forsaking
Foreseeability, 66 Current Legal Probs. 107 (2013).
Contract doctrine classifies these consequential damages as expectation

damages, but they are a consequence of As reliance upon Bs satisfactory
performance. The failure to perform here, unlike in the foregoing dis-
cussion, is not normally the result of the promisor exercising an option
(although, as discussed below, the failure is sometimes deliberate).
When delivery of Hadleys shaft was delayed, Hadley was forced to
shut down his mill.151 Hadleys claim for lost profits arising from the clo-
sure was denied.152 His failure to have another shaft available was the
unusual circumstance cited by Baron Alderson (and, subsequently,
countless others) for rejecting Hadleys claim for the lost profits. 153
Baxendale controlled the likelihood of the bad eventdelayoccurring.
But the consequences of that bad event were, in large part, controlled by
Hadley. That raises the question: To what extent could Hadley run his
business in reliance upon Baxendales performance?
Baron Alderson recognized only one thing Hadley could have done
to avoid the consequenceshold an extra shaft (an input) in inven-
tory.154 But there were many other things he could have done prior to
handing the shaft over to Baxendale. He could have carried a larger
inventory of flour (the output), or he could have recouped the lost out-
put by running the mill at a higher level of output after the shaft had
arrived. (In effect, that entails carrying a larger inventory of productive
capacityanother input.) There is no reason why Hadley had to hold his
inventory of inputs or output at this one locationhe could have diversi-
fied. Any of these actions would have avoided Hadleys loss of profits (or,
at least, substantially reduced them).
The contract-design problem then comes down to this: Should
Baxendale compensate Hadley for the losses he incurred in reliance
upon satisfactory performance, or should Hadley bear the consequences
so that he might be incentivized to control the costs? Some, like
Professor Melvin Eisenberg, argue in favor of the former:
In most cases involving consequential damages it can be
assumed that the buyer has acted prudently during the period
before the contract was made, because reasons of self-regard
will have led him to do so. Under Posners argument, prudence
would require every factory owner to carry the spare parts for a
virtually complete factory, housed alongside his operating fac-
tory. In fact, however, calculations concerning the optimum
supply of spare parts are enormously complex and must reflect
the probability that parts will fail, the cost of waiting for needed
spare parts that are not kept in inventory, the cost of capital
employed in investing in spare parts, the cost of maintaining
spare parts, and the actual or imputed rental costs for storing
151. Hadley, 156 Eng. Rep. at 145; 9 Ex. at 341.

152. Id.
153. Id. at 151; 9 Ex. at 35556.
154. Id.
spare parts. There is little or no reason to believe either that the

mill owner in Hadley v. Baxendale failed to maintain an optimum
supply of spare parts in the period before the crankshaft broke
or that under a less demanding foreseeability standard individ-
uals or firms would generally fail to optimize just because they
might later enter into a contract.155
The evidence seems to favor the latter. Despite the fact that the U.C.C.
and Restatement (Second) both allow for the recovery of consequential
damages that were reasonably foreseeable,156 disclaimers of consequen-
tial damages are ubiquitous.157
The reason is not terribly surprising. If a carrier like Baxendale were
to be liable for the shippers lost profits, and if it could neither price the
risk nor disclaim the liability, then it would in effect be providing manda-
tory insurance to all its customers without the tools insurers customarily
use (copayments, deductibles, monitoring, screening, etc.) to cope with
the inevitable adverse selection and moral hazard. That insurance (and
the extra costs of dealing with customers who make their inventory deci-
sions in reliance on compensation from the carrier if things go awry)
would be a cost of doing business that it would have to cover by charging
higher rates to customers. If the costs to the shippers of self-insuring and
self-protecting were less than the implicit insurance of the carriers, then
the disclaimer would result in savings for shippers as a group.
Buyer forms often do allow for the recovery of consequential
damages. In a battle of the forms, given the knockout rule adopted by
most jurisdictions, consequential damages would be recoverable.158 In
negotiated contracts, my impression is that the disclaimers are common,
consistent with the notion that the buyer is in the best position to protect
its reliance. However, I want to note two situations in which the contract
might provide some protection of the buyers reliance.
In the previous section, I suggested that when the contract is defin-
ing the tradeoff between discretion and reliance, the hostility to the
155. Melvin Aron Eisenberg, The Principle of Hadley v. Baxendale, 80 Calif. L. Rev.
563, 582 (1992) [hereinafter Eisenberg, Principle of Hadley].
156. U.C.C. 2-715 (2012); Restatement (Second) of Contracts 347 (1981). The
U.C.C. does not use the term reasonably foreseeable, but, as Professor Eisenberg notes,
the reason to know standard of 2-715(2) of the UCC lends itself naturally to the
reasonably foreseeable interpretation. Eisenberg, Principle of Hadley, supra note 155, at
613 n.128; see also Farnsworth, Contracts, supra note 46, 12.14 (discussing
unforeseeability as limitation).
157. Scott and Schwartz found that clauses limiting damages to repair and replace-
ment were common. Robert E. Scott & Alan Schwartz, Market Damages, Efficient
Contracting, and the Economic Waste Fallacy, 108 Colum. L. Rev. 1610, 1630 (2008). In
her study of internet contracts, Marotta-Wurgler found that almost every one disclaimed
consequential damages. Florencia Marotta-Wurgler, Some Realities of Online Contracting,
19 Sup. Ct. Econ. Rev. 11, 15 (2011).
158. Not all courts adopting the knockout rule would allow recovery of consequential
damages. E.g., Dresser Indus., Inc. v. Gradall Co., 965 F.2d 1442, 1452 (7th Cir. 1992).
notion that one party to the contract has an option to perform or pay the
consequences is misplaced.159 However, the hostility to the option con-
cept has more bite in cases involving claims for consequential damages.
There is a difference between negligently shipping a shaft and willfully
failing to do so. Although contract law is often characterized as a no-fault
regime, courts have found devices for taking fault into account. As
McCormick wrote over seventy years ago:
Would not our courts enhance the realism of the rules and
make them easier for juries to accept if they gave formal
approval to the tendency, written large upon the actual results
of the cases, to discriminate between the liability for consequen-
tial damages of the wilful and deliberate contract-breaker on
the one hand, and of the party who has failed to carry out his
bargain through inability or mischance? Our rules should sanc-
tion, as our actual practice probably does, the award of conse-
quential damages against one who deliberately and wantonly
breaks faith, regardless of the foreseeability of the loss when the
contract was made. We shall then have completed the process,
begun piece-meal in Hadley v. Baxendale, of borrowing from the
French Civil Code its theory of damages in contract.160
There is considerable dispute as to how to distinguish a willful
breach from any other. Cancellation of an order is deliberate, but few
would find it willful. Corbin was disparaging of the very notion of
willfulness:
The word most commonly used to describe such breaches is
wilful. It is seldom accompanied by any discussion of its mean-
ing or classification of the cases that should fall within it. Its use
indicates a childlike faith in the existence of a plain and obvious
159. See supra Part I.A.

160. Charles T. McCormick, The Contemplation Rule as a Limitation upon Damages
for Breach of Contract, 19 Minn. L. Rev. 497, 51718 (1935); see also Ralph S. Bauer,
Consequential Damages in Contract, 80 U. Pa. L. Rev. 687, 700 (1932) (Aversion and
disgust and hatred excited by a defendants wilful breach of contract, and pity aroused by
the predicament in which another defendant is placed by his honest and unintentional
breach of contract, have swayed judges, as well as juries, in the actual administration of the
law.). Fuller and Perdue note that fault is taken into account indirectly by manipulating
the foreseeability doctrine. Fuller & Perdue, Reliance 1, supra note 1, at 8586. For more
recent arguments for recognizing fault and willfulness, see George M. Cohen, The Fault
Lines in Contract Damages, 80 Va. L. Rev. 1225, 1226 (1994) (arguing for different
measures of damages depending in part on fault); Melvin Aron Eisenberg, The Role of
Fault in Contract Law: Unconscionability, Unexpected Circumstances, Interpretation,
Mistake, and Nonperformance, 107 Mich. L. Rev. 1413, 1414 (2009) (As a normative
matter, fault should be a building block of contract law.); Steve Thel & Peter Siegelman,
Willfulness Versus Expectation: A Promisor-Based Defense of Willful Breach Doctrine, 107
Mich. L. Rev. 1517, 1518 (2009) (arguing willfulness identifies easily avoided breaches that
courts rightly try to deter); Robert A. Hillman, The Future of Fault in Contract Law 2, 4
(Cornell Law Sch. Legal Studies Working Paper No. 12-34, 2012), available at
http://ssrn.com/abstract_id=2121374 (on file with the Columbia Law Review) (noting
fault plays an important role in contract law today and increase in opportunities for
advantage-taking suggest an even larger role for fault . . . in the future).
line between the good and the bad, between unfortunate virtue
and unforgivable sin.161
Contrast this with his enthusiastic approval of good faith, which surely
suffers from the same flaw.162 Notwithstanding the difficulties, courts do
make that distinction and, more importantly, the parties themselves
often do so in their contracts. Even if they disclaim liability for conse-
quential damages, they can include a significant exceptionthe
disclaimer will not apply if the breach were due to gross negligence or
willful behavior.163 They might not have any idea about what they mean
by willful; rather than spelling it out, they are content to defer the defini-
tion to the ex post determination by courts.164 The disposition of Hexion
Specialty Chemicals, Inc. v. Huntsman Corp., 165 discussed in Part I.A.4,
provides one illustration. The Chancery Court distinguished Hexions
knowing and intentional breach of [a] covenant and allowed for liabil-
ity substantially greater than if the transaction had been terminated for
an acceptable reason.166
In a Hadley-type scenario, a deliberate deviation by a carrier to pick
up a more valuable shipment could result in liability for the shippers
consequential damages. Of course, that can be contracted overthe
seller could maintain the flexibility to deviate.167 The shipper might be
161. 12 Joseph M. Perillo, Corbin on Contracts 62.2 (rev. ed. 2012). For a more
recent and more measured, skeptical take on willfulness, see Richard Craswell, When Is a
Willful Breach Willful? The Link Between Definitions and Damages, 107 Mich. L. Rev.
1501, 150105 (2009) (noting labels like willfully are not self-defining and arguing
specific definition controls magnitude of damages).
162. 6 Peter Linzer, Corbin on Contracts 26.9 (Joseph Perillo ed., rev. ed. 2010)
[hereinafter Linzer, Corbin] (Good faith is a vague and shifting concept, but so is justice.
That a concept cannot be formalized into a tight matrix does not make it wrong. It makes
it consistent with the way humans behave . . . .).
163. One of the standard forms of the Federation of Oils, Seeds and Fats Associations
has an interesting variant on the effect of fault on damages: If the arbitrators consider the
circumstances of the default justify it they may, at their absolute discretion, award damages
on a different quantity and/or award additional damages. Novasen SA v. Alimenta SA,
[2013] EWHC (Comm) 345, [3].
164. See Scott & Triantis, Anticipating Litigation, supra note 6, at 816 (providing
powerful analysis of tradeoff between defining terms at front end and deferring definition
to a third partycourt or arbitratorat back end).
165. 965 A.2d 715 (Del. Ch. 2008).
166. Id. at 724, 75657.
167. Ocean World Lines Bill of Lading Terms & Conditions provides an illustration
of that flexibility:
The Carrier does not undertake that the Goods will be transported from or
loaded at the place of receiving or loading or will arrive at the place of
discharge, destination or transshipment aboard any particular vessel or other
conveyance or at any particular date or time or to meet any particular market or
in time for any particular use. Scheduled or advertised departure and arrival
times are only expected times and may be advanced or delayed if the Carrier or
any Connecting Carrier shall find it necessary, prudent or convenient. In no
event shall the Carrier be liable for consequential or other damages for delay in
content to give the carrier that option in exchange for a lower price.
Suppliers in many contexts do offer interruptible service at a reduced
price.
The willfulness exception can be rationalized in terms of the buyers
reliance. For innocent mistakes by the seller, the onus is on the buyer to
protect itself. Buyers can make their decisions relying on the sellers
normal behavior, even if that behavior results in a seller breach. Behav-
ior by the seller that substantially increases the likelihood of failure,
however, would be outside the buyers reasonable expectationsbuyers
need not self-protect against that. However, recognizing a fault-based
exception can expose the seller to juridical risk. Given the difficulty in
defining willful behavior and the risk that a factfinder would define a
garden-variety failure as willful, parties might be reluctant to include a
willfulness exception.168
The second exception mirrors the second prong of the Hadley rule:
damages as may reasonably be supposed to have been in the contempla-
tion of both parties, at the time when they made the contract, as the
probable result of the breach.169 It is not the mere contemplation that
matters. The sellers have knowledge of the buyers vulnerability, they
know that the buyer is relying on their performance, and they accept the
risk that their failure would result in substantial damages. Contracts
between suppliers and automobile manufacturers make clear the buyers
intended use170 and assess sellers for at least some of the costs if there is a
the scheduled departures or arrivals of the vessel or other conveyance

transporting the Goods or for any other matter.
Bill of Lading Terms & Conditions, Ocean World Lines, http://www.oceanworldlines.com
/BL-terms-conditions.aspx (on file with the Columbia Law Review) (last visited Mar. 24,
2014).
168. As an example of the juridical risk, consider this disclaimer by a grain elevator
operator:
The elevator management, in its sole discretion, may alter the turn of
vessels to be loaded [a] when, in its judgment, it is in the best interest of the
elevator operations; [b] when there is urgent need to receive or load to a ship a
particular grade and kind of grain; [c] to facilitate the berth conditions; or [d]
whenever the elevator decides that there is nonavailability in the elevator of
adequate grade, quantity, or quality of grain to be shipped or loaded on the
vessel without delaying the vessel itself, or delay, prevent, or obstruct the normal
elevator activity.
Romano Pagnan & F.LLI v. Miss. River Grain Elevator, Inc., 700 F.2d 149, 151 n.2 (5th Cir.
1983). The contract further states: The elevator shall not be liable for demurrage,
damages for delay or loss of dispatch time incurred by any vessel or charterer for any cause
other than wilful or gross negligent acts of the elevator management. Id. at 151 n.3. A
jury, and the court of appeals, found a fairly minor delay to be willful. Id. at 150.
169. Hadley v. Baxendale, (1854) 156 Eng. Rep. 145 (Ex.) 151; 9 Ex. 341, 354.
170. See GM Standard Form, supra note 107 (Seller acknowledges that Seller knows
of Buyers [intended use] and expressly warrants that all goods covered by this order
which have been selected, designated, manufactured, or assembled by Seller based upon
Buyers stated use, will be fit and sufficient for the particular purposes intended by
Buyer.).
breach of warranty (to the consumer) or a product recall. Fords stand-

ard form, for example, says: At its option, the Buyer may debit the
Supplier for up to 50% of the Actual Recall Costs . . . if the Buyer has
made a good faith determination that the Supplier is likely to be liable
for some portion of the total costs . . . .171
The multiyear supply contract between John Deere and Stanadyne
makes liability for consequential damages contingent on both knowledge
and fault.172 The wording reverses the usual pattern of saying no liability
unless the seller passes some culpability hurdle. Instead the seller would
be liable unless it is without fault:
STANADYNE CORPORATION acknowledges that DEERE
requires on-time delivery in order to operate its plants. The par-
ties further acknowledge that the precise amount of damages
which DEERE would sustain in the event STANADYNE
CORPORATION were to fail to make timely or conforming
deliveries of Parts would be difficult to determine. Therefore,
the parties agree that STANADYNE CORPORATION shall be
responsible for any and all damages resulting from
STANADYNE CORPORATIONs failure to make timely or con-
forming deliveries of Parts, including, but not limited to, mutu-
ally agreed upon costs DEERE incurs for the correction of Parts
with quality problems and mutually agreed upon costs DEERE
incurs in connection with DEEREs machining and/or assembly
line downtime . . . . STANADYNE CORPORATION shall not be
responsible for the above damages if such out-of-order (late)
delivery or non-delivery results from a cause beyond
STANADYNE CORPORATIONs reasonable control without
fault or negligence, provided that STANADYNE
CORPORATION has informed DEERE as soon as practical of
the problem . . . .173
171. Ben-Shahar & White, supra note 108, at 959 n.24 (quoting Ford Motor Co.,
Production Purchasing Global Terms and Conditions 23.06 (Jan. 2004)). The Toyota
form goes further:
Upon the occurrence of a Recall or Products Liability Situation (Recall and
Products Liability Situation are collectively referred to as a Reimbursement
Event), where one of the potential causes for the Reimbursement Event is
determined in Toyota Partys reasonable judgment to be attributable to
Supplier, Supplier and Toyota party agree to negotiate in good faith to (i)
reasonably allocate the cost of complying with or contesting any reimbursement
event and (ii) determine TMMNAs remedies under this Section 5.4 which may
include actual, consequential and incidental damages (including, without
limitation, attorneys fees and administrative costs and expenses) arising out of,
resulting from or related to any such Reimbursement Event.
Toyota T&C, supra note 108, cl. 5.4(a).
172. Long Term Agreement Between Deere & Co. and Stanadyne Corp. (Nov. 1,
2001), available at http://www.sec.gov/Archives/edgar/containers/fix023/1053439/0001
19312507182449/dex1011.htm (on file with the Columbia Law Review).
173. Id. cl. XV.A.
I must reiterate that these remain exceptions to the basic point.

When parties design their contracts, rarely will they protect the promi-
sees reliance on the promisors successful performance. The promisees
ability to control the adverse consequences usually would result in con-
sequential damages being disclaimed.
III. RELIANCE AND RESTITUTION WHEN PERFORMANCE IS EXCUSED
Performance of a contract could be excused under various doc-

trinesimpossibility, impracticability, and frustrationor under the
terms of the contractforce majeure or MAC. Prior to the occurrence of
the event that excused performance, the parties might have incurred
costs in reliance on the contract. A buyer who had made a partial
payment may want restitution; a seller who had advertised an event or
partially completed performance may want compensation for its reliance
costs. The legal analysis of both sets of claims has been muddled by
notions of fairness and justice (or, as the commentators prefer, avoiding
injustice).174 Thus, if performance is excused, the Restatement (Second)
of Contracts says that either party may have a claim for relief including
restitution . . . . [I]f those rules . . . will not avoid injustice, the court may
grant relief on such terms as justice requires including protection of the
parties reliance interests.175 Fuller and Perdue argue that the equitable
resolution entailed recognition of the restitution and reliance interests:
[T]he most satisfactory solution of the difficulty may well be to relieve
the promisor from his duty, at the price, not simply of returning benefits,
but of making good the other partys losses through reliance on the
contract.176
174 . Much of the following discussion is based on Victor P. Goldberg, After

Frustration: Three Cheers for Chandler v. Webster, 68 Wash. & Lee L. Rev. 1133 (2011)
[hereinafter Goldberg, After Frustration].
175. Restatement (Second) of Contracts 272 (1981). In England, the Law Reform
(Frustrated Contracts) Act provides for restitution of money paid before the discharging
event subject to a key proviso. The repayment could be offset, in whole or in part, if that
party had incurred reliance expenditures if the court considers [it] just, having regard to
all the circumstances of the case. Frustrated Contracts Act, supra note 14, 1.
176. L.L. Fuller & William R. Perdue, Jr., The Reliance Interest in Contract Damages:
2, 46 Yale L.J. 373, 380 (1937). The full context of their argument is as follows:
In such a field, where no technical rules serve to obstruct an insight into the
purposes underlying contract law, it would seem inevitable that the cases would
reveal a distinction between the three interests which have been described in this
article. Such a distinction has definitely been taken in America (though
apparently not in England) between the expectation and restitution interests.
Where a contract has become impossible of performance, or its object is
frustrated, it has been recognized that the most equitable adjustment of the
situation may call for relieving the party from liability for future performance
(expectation interest denied), and at the same time imposing on him a duty to
return benefits received under the contract (restitution interest protected).
But in this field, where borderline cases are a normal phenomenon, it
would seem that the reliance interest should also play an important role. Where
In the first Anglo-American case recognizing the impossibility

defense, Taylor v. Caldwell, a venue was destroyed before the contracted-
for event (a concert series) was to take place.177 The promoter sued for
his reliance costs (58), 178 the costs incurred in preparation for the
intended concert series.179 The claim was denied on the grounds that
there had been no breach; there was no focus on whether there could be
compensation for reliance when performance had been excused,
although a negative answer was implicit in the decision. 180 In the
Coronation Cases, arising from the postponement of the coronation
procession following King Edwards appendicitis attack, the House of
Lords considered the question and left the losses where they fell.181 The
renter, under Chandler v. Webster, was responsible for any payments that
had been made or were due before the procession was postponed.182
There would be no restitution.183
This did not sit well with many. Contrasting the English rule with
that of Scotland, the House of Lords noted in high legal quarters a feel-
ing both of uneasiness and of disrelish as to the English rule, which
works well enough among tricksters, gamblers and thieves.184 After dec-
ades of critical comment, the House of Lords reversed itself in Fibrosa
Spolka Akcyjna v. Fairbairn Lawson Combe Barbour, Ltd.185 A British manu-
facturer agreed to manufacture a flax-hackling machine for a Polish
buyer for delivery in three to four months. The buyer paid about a third
of the contract price when the order was placed, and the manufacturer
the court is in doubt whether the excuse should be permitted at all, the most
satisfactory solution of the difficulty may well be to relieve the promisor from his
duty, at the price, not simply of returning benefits, but of making good the other
partys losses through reliance on the contract. In Germany, the Civil Code
expressly recognizes the usefulness of the reliance interest as a means of
accomplishing the most equitable allocation of the risks involved in impossibility.
Id. at 37980 (footnotes omitted).
177. (1863) 122 Eng. Rep. 309 (K.B.) 312, 315; 3 B. & S. 826, 832, 840.
178. Victor P. Goldberg, Excuse Doctrine: The Eisenberg Uncertainty Principle, 2 J.
Legal Analysis 359, 366 (2010) [hereinafter Goldberg, Excuse Doctrine]. Per capita
annual income in England at the time was only around 40. Gregory Clark, Average
Earnings and Retail Prices, UK, 12092010, at 2 (Oct. 30, 2011) (unpublished
manuscript), available at http://www.measuringworth.com/datasets/ukearncpi/earnstudy
new.pdf (on file with the Columbia Law Review).
179. Taylor, 122 Eng. Rep. at 310; 3 B. & S. at 827.
180. Id. at 315; 3 B. & S. at 840.
181. See R.G. McElroy & Glanville Williams, The Coronation CasesI, 4 Mod. L. Rev.
241, 245 & nn.1821 (1941) (citing, categorizing, and discussing cases).
182. [1904] 1 K.B. 493 (C.A.) at 497 (Eng.).
183. Id.
184. Cantiare San Rocco v. Clyde Shipbuilding & Engg Co., [1924] A.C. 226 (H.L.)
25859 (appeal taken from Scot.).
185. [1943] A.C. 32 (H.L.) 7172 (appeal taken from Eng.). Scottish law had rejected
Chandler earlier. It attempted to restore the precontract situation, requiring the refund of
money prepaid and also compensating the seller for at least some of its reliance costs.
Cantiare San Rocco, [1924] A.C. at 229.
started production. 186 Unfortunately, the Germans decided to invade

Poland before the machine was finished.187 The House of Lords ruled
that the contract had been frustrated and then overruled Chandler v.
Webster.188 The seller would have to return the prepayment.189 It did not,
however, require any compensation to the seller for costs it might have
incurred in reliance on the contract.190 It might be unfortunate, said
Lord Chancellor Simon, if the seller had relied, but, absent specific
language, the reliance would not be protected:
While this result obviates the harshness with which the
previous view in some instances treated the party who had made
a prepayment, it cannot be regarded as dealing fairly between
the parties in all cases, and must sometimes have the result of
leaving the recipient who has to return the money at a grave
disadvantage. He may have incurred expenses in connexion
with the partial carrying out of the contract which are equiva-
lent, or more than equivalent, to the money which he prudently
stipulated should be prepaid, but which he now has to return
for reasons which are no fault of his. He may have to repay the
money, though he has executed almost the whole of the con-
tractual work, which will be left on his hands.191
New legislation, he said, would be required. Parliament responded
shortly thereafter by passing the Law Reform (Frustrated Contracts)
Act,192 which allowed for compensation for reasonable reliance. About
the best thing that can be said about the Act is that it is virtually never
used. Instead of leaving the parties where they were when the contract
was excused (the Chandler rule), the Act would give the buyer restitution
of any prepayment, and the seller, who incurred reliance costs, might be
able to offset some or all of that if a court thought it would be just to do
so. What does that standard look like? In one of the rare cases applying
the Act, Lord Justice Lawton said that the law grants the trial court
considerable discretion: What is just is what the trial judge thinks is just.
That being so, an appellate court is not entitled to interfere with his deci-
sion unless it is so plainly wrong that it cannot be just.193
As noted, the American rule is similar. In neither case is there any
thought about why the contracts might have structured the performance
and payment obligations as they had. The phasing of payment and per-
formance is not merely a matter of whim; it is a decision variable of the
parties. The contract could determine whether either party should be
186. Fibrosa Spolka Akcyjna, [1943] A.C. at 34.

187. Id.
188. Id. at 40, 49, 53.
189. Id. at 49.
190. Id. at 5859.
191. Id. at 49.
192. Frustrated Contracts Act, supra note 14.
193. B.P. Exploration Co. (Libya) v. Hunt (No. 2), [1981] 1 W.L.R. 232 (C.A.) at 238
(Eng.).
compensated after performance was excused; or it could simply leave the

parties as they were at the time the excusing event occurred. There are
many reasons why a buyer might make some payments before a project is
completed and why it might choose to make some or all of those pay-
ments nonrefundable. The payment might be for an option or com-
pensation for a products customization; alternatively, the payment may
serve to provide working capital to the seller or to assure the seller that,
after it had completed most of the performance, the buyer would not
have the incentive to renegotiate (and, in doing so, take advantage of
sunk costs).194
The Restatement (Second)s illustration of the problem is instruc-
tive. A (a contractor) is working on an existing structure owned by B, but
before completion of the job, the structure is destroyed by fire and the
contractor is excused.
A contracts with B to shingle the roof of Bs house for
$5,000, payable as the work progresses. After A has spent $2,000
doing part of the work and has been paid $1,800, much of the
house including the roof is destroyed by fire without his fault,
and the duties of performance of both A and B are dis-
charged . . . . The work done before the fire increased the mar-
ket price and the insurable value of the house by $1,500. A is
entitled to restitution of $1,500 from B and B is entitled to resti-
tution of $1,800 from A . . . .
. . . [T]he fire also destroyed shingles that had cost A $500
and that were piled near the house for the rest of the work. A is
not entitled to restitution of this loss from B. The court may,
however, take this loss into consideration in deciding whether
to allow A restitution of $1,500 or $2,000.195
No reasons are given for the numbers. Why has B paid more than
the benefit it had thus far received and less than As costs? Why treat
the shingles piled near the house differently? If one cares about reliance,
then there is no reason to single these out. Why care at all about the
benefit B received since after the fire there were no benefits? Why go
through the effort of ascertaining the value added pre-fire and the
contractors costs (most of which would likely be labor costs)? Accidents
during construction projects are hardly unexpected. One would expect
parties to anticipate these problems in their contract design. And, in fact,
they do. What they do is ignore both reliance and restitution.196 The con-
tract is terminated, and neither party has to compensate the other. The
close link between progress payments and performance will mean that
194. See generally Goldberg, After Frustration, supra note 174, at 1146 (listing
reasons why buyer might prepay some, or all, of purchase price).
195. Restatement (Second) of Contracts 377 illus. 45 (1981).
196. See Goldberg, After Frustration, supra note 174, at 116567 (giving real-world
examples of parties contracting around Restatement (Second) rule, often through
insurance requirements).
leaving the losses where they fall will generally work well. The best thing
that can be said for the doctrinal solution is that it is irrelevant.197
I do not mean to suggest that, in all instances in which performance
would be excused, parties would never deviate from the leave the losses
where they fall resolution. But they would deviate as a matter of plan-
ning, not as a matter of justice.
While the language of excuse cases (and the scholarly literature)
often suggests that the parties did not, or could not, have planned for the
specific event, it misses the point. The Kings appendicitis (the
Coronation Cases) might have been beyond their contemplation, but the
possibility that the procession might have been postponed for any reason
was not. In fact, there was a very active insurance market, and a number
of contracts did explicitly recognize the possibility of postponement.198
More generally, as Triantis has argued, specific risks can be allocated as
part of a more broadly defined risk.199 Planners could, if they so desired,
differentiate between categories of risk when determining whether there
should be some compensation for restitution or reliance. So, for exam-
ple, a Rod Stewart contract distinguished between two categories of
excuse.200 If the performance could not be rendered because of the usual
acts of God (floods, fires, riots, strikes, etc.), the performance would be
rescheduled; if he were ill or incapacitated, the show would be cancelled
and he would have to refund any prepayments.201
IV. RELIANCE AND REVOCATION
The interplay of reliance and injustice shows up in other sections of

the Restatement (Second), notably in section 90(1) (making a promise
absent consideration enforceable)202 and section 87(2) (making an offer
irrevocable). 203 Grant Gilmores prediction that promissory estoppel
197. Because the contracts typically resolve the problem, most of the case law
providing the basis for the Restatement (Second) dates back to the First World War. See
id. at 116265 (discussing early-twentieth-century American case law dealing with
frustration doctrine).
198. See Goldberg, Excuse Doctrine, supra note 178, at 36266 (discussing insurance
market for contract postponement in Coronation Cases).
199. George G. Triantis, Contractual Allocations of Unknown Risks: A Critique of the
Doctrine of Commercial Impracticability, 42 U. Toronto L.J. 450, 46468 (1992).
200. See Goldberg, Excuse Doctrine, supra note 178, at 36869 (discussing force
majeure clause in Rod Stewarts contract).
201. Id. at 10.
202. Restatement (Second) of Contracts 90(1) (1981) (A promise which the
promisor should reasonably expect to induce action or forbearance . . . and which does
induce such action or forbearance is binding if injustice can be avoided only by
enforcement of the promise. The remedy granted for breach may be limited as justice
requires.).
203. Id. 87(2) (An offer which the offeror should reasonably expect to induce
action or forbearance of a substantial character on the part of the offeree before
would swallow up the consideration doctrine has not come to passquite

the opposite.204 It remains a bit player in determining the enforceability
of promises. Justice Roger Traynors innovationthe irrevocable offer
has fared even worse. Aside from the context of subcontractor bids in
public construction projects, it has had virtually no impact. My concern
here is not with the impact of the decision. I have considered that else-
where.205 Rather, I want to focus on the cases within the Drennan ambit to
illustrate how slippery the reliance concept is when courts try to imple-
ment it.
Reliance can mean different things to different people (and courts);
in practice it means so many different things that it is not clear that it
means anything at all.206 In classic contract law, an offeror is free to
revoke an offer before acceptance and a counteroffer is a rejection of an
offer. Traynors innovation was to use reliance to make the offer irrevo-
cable. In Drennan, after the general contractor (GC) had been named
the successful low bidder, the subcontractor (sub) informed him that it
had erred in preparing its bid (an offer), and it was therefore withdraw-
ing it.207 The GC completed the project with a new sub at a higher price
and then sued the original sub for the difference.208 [T]he question,
wrote Traynor, is squarely presented: Did plaintiffs reliance make
defendants offer irrevocable?209 Given . . . [that the GC] is bound by
his own bid, Traynor stated, it is only fair that [the GC] should have at
least an opportunity to accept [the subs] bid after the general contract
has been awarded to him.210 However, he added a qualification: It
bears noting that a general contractor is not free to delay acceptance
after he has been awarded the general contract in the hope of getting a
better price. Nor can he reopen bargaining with the subcontractor and at
the same time claim a continuing right to accept the original offer.211
Doing so, he suggested, would destroy reliance.212
acceptance and which does induce such action or forbearance is binding as an option
contract to the extent necessary to avoid injustice.).
204. See Grant Gilmore, The Death of Contract 68 (Ronald K.L. Collins ed., 1995)
(The one thing that is clear is that these two contradictory propositions cannot live
comfortably together: in the end one must swallow the other up.).
205. See Victor P. Goldberg, Traynor (Drennan) Versus Hand (Baird): Much Ado
About (Almost) Nothing, 3 J. Legal Analysis 539, 57885 (2011) [hereinafter Goldberg,
Drennan] (discussing limited impact of irrevocable offer outside public construction
context).
206. Of course, in another context Justice Traynor celebrated the indeterminacy of
language. Pac. Gas & Elec. Co. v. G.W. Thomas Drayage & Rigging Co., 442 P.2d 641, 643
46 (Cal. 1968).
207. Drennan v. Star Paving Co., 333 P.2d 757, 75859 (Cal. 1958) (en banc).
208. Id. at 759.
209. Id.
210. Id. at 760.
211. Id.
212. Id.
The GCs reliance, therefore, is a question of fact. How could the

court ascertain whether the GC had in fact relied upon the subs bid?
Traynors opinion set out the criteria. The GC could not be found to
have reasonably relied if: (1) the GC should have known the bid was in
error; (2) the GC proposed new terms (the counteroffer); or (3) GCs in
general, or this specific GC, were known to shop the bid, and/or the GC
shopped the bid in this instance.213 Any of these might be enough to
defeat reliance, but implementation, especially of the last two, left much
to be desired.
Consider first the counteroffer. Subs usually submit their bid to a
number of general contractors. In most jurisdictions, the GC must
include the name of the sub it used in preparing its proposal.214 After
being named the winning bidder, the GC would send a written contract
to that sub. If the terms of that contract were deemed a material altera-
tion of the subs offer, then it could be treated as a counteroffer. Since in
most instances the subs offer consisted of a single number (with no
stated terms), it should not be surprising that the terms of the GCs form
contract would differ. If the court found the GCs form terms materially
different, it would have to conclude that the GC would no longer be rely-
ing upon the subs offer, so the sub would be free to reject the counter-
offer. That is, the offer would be revocable.
In some instances subs raised this argument in defense, and on occa-
sion they succeeded.215 The courts listed the nonconforming clauses and
labeled them as either material or nonmaterial, although in most inst-
ances it is hard to tell how the court came to the conclusion that it did.
For example, in one of the cases holding for the sub, the court recog-
nized two clauses (among the seven) that made the GCs written contract
a counteroffer: [T]he sub-contract prohibited the sub-contractor from
continuing to employ any person deemed by the owner, architect or con-
tractor to be a nuisance or a detriment to the job . . . [and] the
subcontract authorized the architect to discharge any workman committ-
ing a nuisance upon certain parts of the premises. 216 It is hard to
imagine that these would be dealbreakers. Of course, the materiality of
the contested terms in the cases raising the defense had nothing to do
with the subs rejectionit was simply a device for getting out of a bad
deal. Nonetheless, the argument succeeded in negating reliance in
almost half the cases in which the courts considered it.217
Traynors position on the role of bid shopping in negating reliance
seems clear: A GC could not reopen bargaining with the subcontractor
213. Id. at 76061.

214. Most states require the listing of a sub if its bid exceeds a threshold. See
Goldberg, Drennan, supra note 205, at 57075 (discussing state listing statutes).
215. See id. at 57780 (discussing cases where subs have succeeded).
216. Hedden v. Lupinsky, 176 A.2d 406, 407 (Pa. 1962).
217. Goldberg, Drennan, supra note 205, at 559.
and at the same time claim a continuing right to accept the original
offer.218 If bid shopping were common in this particular market, could a
GC rely on a subs offer? In a Minnesota case, the court noted that the
sub contended bid shopping and bid chopping are so common to
the Twin Cities area construction industry that [contractors] do not
expect to be bound by prices submitted by the subcontractors . . . and
that defendant was therefore not bound on its bid . . . because further
and final negotiations would take place at a later time.219 However, since
there was no evidence that the GC had shopped this particular bid, the
court found for the GC, holding that its reliance was reasonable.220 This
is a really peculiar argument. In effect the court is saying that no one in
this market relies on the prices quoted by subs, but that in this one case
the GC did and the reliance was reasonable.
Another case added an odd twist to the reliance argument. In Saliba-
Kringlen Corp. v. Allen Engineering Co., the GC, claimed the court, did not
rely on the offered price per se; rather, it relied on the subs offer setting
a ceiling that would allow it to freely bargain for a better price from this
sub or its competitors without having the offer lapse.221 That is, the GC
could shop the bid at will, so long as it ended up with a price at or below
the subs bid. If it failed to do better, it could still hold the sub to its bid.
This is hardly what Traynor had in mind. This case is an outlier, although
one economic analysis of Drennan does treat the subs irrevocable offer as
a cap.222 The decision does, however, indicate the broad discretion that
courts have in applying the ill-defined reliance standard to the question.
If the subs bid was treated as an irrevocable offer, the GC would
have a valuable option. The value increases with the length of time and
the variance of the subs costs (in particular, its opportunity costs). If the
expected value of the option to the GC was greater than the expected
cost to the sub of providing it, the parties would have an incentive to
agree to make the option irrevocable for some period of time. This does
not mean that the GC would have to negotiate the revocability issue with
each potential sub; all that would be necessary is that the GC set out the
criteria for revoking a bid in the bidding documents (and a presumption
that courts would enforce the terms of the bidding documents).223 The
218. Drennan, 333 P.2d at 760.

219. Constructors Supply Co. v. Bostrom Sheet Metal Works, Inc., 190 N.W.2d 71, 76
(Minn. 1971).
220. Id. at 77.
221. 92 Cal. Rptr. 799, 80209 (Ct. App. 1971).
222. See Ofer Grosskopf & Barak Medina, Rationalizing Drennan: On Irrevocable
Offers, Bid Shopping and Binding Range, 3 Rev. L. & Econ. 231, 238 (2007) (assuming
GCs are free to request additional bids after winning general contract).
223. For an example of a court ignoring the GCs requirement that bids be held open
for a period, see Fletcher-Harlee Corp. v. Pote Concrete Contractors, Inc., 482 F.3d 247,
24952 (3d Cir. 2007).
Drennan rule really gets it backwards.224 Instead of having reliance deter-

mine the level of irrevocability, the law should allow the GC to determine
the level of irrevocability it needs to protect its reliance.
The reliance, according to Justice Traynor, was not symmetrical.225
Subs could not claim that they relied on the GC.226 Why? Because the sub
typically submits a bid to a number of GCs, it therefore relies on none of
them. Subs have invariably lost, although one court used the asymmetry
to deny the GCs reliance claim and stated:
Allowing a cause of action based on promissory estoppel in
construction bidding also creates the potential for injustice. It
forces the subcontractor to be bound if the general contractor
uses his bid, even though the general contractor is not obli-
gated to award the job to that subcontractor. The general con-
tractor is still free to shop around between the time he receives
the subcontractors bid and the time he needs the goods or ser-
vices, to see if he can obtain them at a lower price.227
When preparing its bid, whether the sub relied on this GC or on all
the GCs to whom it submitted its bid is really beside the point. The GCs
commitment to the sub named in its bid could be spelled out in the bid-
ding documents. Apparently GCs are reluctant to make such a commit-
ment. As a rule, the sub does not get that protection by contract. Indeed,
subs have attempted to overcome the asymmetry by group action, either
through legislation or through private organizations like bid deposito-
ries.228 Their efforts have only had limited success.229 The relevant point
is that whether or not a sub relied upon a GC says nothing about the
extent to which a GC should be bound to use a sub.
V. RELIANCE AND INFORMATION

In complex contracts, like a corporate-acquisition contract, the par-
ties face two significant problems. First, there is the adverse-selection
problem; the buyer needs information regarding the quality of what it is
buying. (If the deal were being financed in part by stock, the seller would
need information regarding the quality of the buyer as well.) Second,
there can be a temporal gap between execution of the contract and the
closing, and a lot can happen in that period that would affect the value
224. See supra notes 207218 and accompanying text (discussing Drennan decision).
225. S. Cal. Acoustics Co. v. C.V. Holder, Inc., 456 P.2d 975, 979 (Cal. 1969) (en
banc).
226. Id.
227 . Home Elec. Co. of Lenoir, Inc. v. Hall & Underdown Heating & Air
Conditioning Co., 358 S.E.2d 539, 542 (N.C. Ct. App. 1987), affd, 366 S.E.2d 441 (N.C.
1988).
228. See Goldberg, Drennan, supra note 205, at 57076 (discussing two potential
remedies for disadvantaged subcontractors).
229. Id. at 576 (Indeed, even with many of those restrictions in place, postbid
negotiation appears to be common.).
of the seller. Reliance is implicated in both problems. Information is

costly to produce, and in many instances, the seller will be the best
source of information regarding its quality. With little information, the
buyer fears that it is buying a lemon and would offer a price that refl-
ected that fear.230 The seller has an incentive to provide information that
would reassure the buyer; it would provide a package of representations
and warranties that would fill in the details. The contract would have to
specify on which of the sellers statements the buyer would rely and how
that reliance would be protected.231
The acquisition document will delineate the information on which
the buyer should and should not rely. Integration clauses and the parol
evidence rule restrict, to some extent, the ability of courts to look outside
the written agreement. While there is much hostility to the parol
evidence rule amongst the professoriate, my understanding is that, espe-
cially for sophisticated parties, the courts in most jurisdictions generally
respect these clauses. Professor Peter Linzer, one of the scholars hostile
to the rule, reluctantly concludes in his update of Corbin on Contracts that
courts generally respect the parol evidence rule.232 Professors Schwartz
and Scott, decidedly less hostile, likewise found that the rule was alive
and well.233 I do not want to be drawn into that debate. My concern here
is with two subtopics: no-reliance clauses and remedies for breach of a
warranty. A central concern in both is recognizing that litigation is
expensive and imperfect. The leading New York case, Danann Realty
Corp. v. Harris,234 does not quite fit the bill. It involves a contract for
purchase of a lease of a buildinga simpler transaction in which it is at
least plausible that the buyer was not aware of the no-reliance lan-
guage.235 Nonetheless, it is instructive.
The buyer in Danann contended that it was induced to enter into
the sale by false oral representations about operating expenses and prof-
its that it would derive from the investment.236 The contract included in
the merger clause a disclaimer of any representations regarding a litany
of items including the operating expenses and profits, except as herein
specifically set forth [in the agreement].237 The majority held that if this
230. George A. Akerlof, The Market for Lemons: Quality Uncertainty and the
Market Mechanism, 84 Q.J. of Econ. 488, 488500 (1970).
231. The discussion will be confined to the acquisition of a private company or the
division of a public company to avoid the additional complication of the fiduciary duties of
the sellers board.
232. 6 Linzer, Corbin, supra note 162, 25.1.4.
233. Alan Schwartz & Robert E. Scott, Contract Interpretation Redux, 119 Yale L.J.
926, 95861 (2010).
234. 157 N.E.2d 597 (N.Y. 1959).
235. Id. at 598.
236. Id.
237. Id.
were a general and vague merger clause, it would be ineffective to

exclude parol evidence to show fraud in inducing the contract.238 But:
Here . . . plaintiff has in the plainest language announced
and stipulated that it is not relying on any representations as to
the very matter as to which it now claims it was defrauded. Such
a specific disclaimer destroys the allegations in plaintiffs com-
plaint that the agreement was executed in reliance upon these
contrary oral representations.239
Since the buyer should have read and understood the contract, the court
continues, it would be unrealistic to ascribe to plaintiffs officers such
incompetence that they did not understand what they read and
signed.240 Implicit in this argument is the notion that the clause was tai-
lored to this particular transaction. The dissent denies this, noting that
the clause was boilerplate.241 If a generic merger clause was problematic,
then why should a boilerplate clause do any better? The dissents primary
argument does not rely on the fact that the clause was a standard clause.
Even if the clause had been bargained over, the buyer was entitled to its
day in court because fraud vitiates every contract and every clause in
it.242
The case is complicated a bit by the fact that the disputed clause was
not bargained over. The buyer should have had legal advice, and the
advisor should have had knowledge of the language. But one could at
least argue that the majoritys distinction between this clause and a
generic merger clause failed. The duty to read should apply equally to
both or to neither. The more significant issue concerns no-reliance
clauses that are specifically negotiated. The Danann rule has been
applied to the more complex transactions in the two primary commercial
jurisdictions, New York and Delaware.243
In a decision in which he recognized a no-reliance clause, Judge
Posner distinguished contracts between sophisticated parties:
In the trade, no-reliance clauses are called big boy
clauses (as in were big boys and can look after ourselves). But
238. Id.
239. Id. at 599.
240. Id.
241. Id. at 602 (Fuld, J., dissenting).
242. Id. at 603.
243. For application of the rule in New York, see Grumman Allied Indus., Inc. v. Rohr
Indus., Inc., 748 F.2d 729, 735 (2d Cir. 1984). For its application in Delaware, see Abry
Partners V, L.P. v. F & W Acquisition, L.L.C., 891 A.2d 1032, 1058 & n.57 (Del. Ch. 2006).
But in other jurisdictions, the Danaan rule has not been applied in similar contexts. See
Allen Blair, A Matter of Trust: Should No-Reliance Clauses Bar Claims for Fraudulent
Inducement of Contract?, 92 Marq. L. Rev. 423, 44550 (2009) (explaining theoretical
objections to Danaan rule); Kevin Davis, Licensing Lies: Merger Clauses, the Parol
Evidence Rule and Pre-Contractual Misrepresentations, 33 Val. U. L. Rev. 485, 51314
(1999) (discussing limited instances of enforcement of nonreliance clauses under Danaan
rule).
if someone who is not a big boyindeed is not even represented

by counselsigns a big-boy clause, there can be a problem, and
this has led some courts to require, before such a clause can be
enforced, an inquiry into the circumstances of its negotiation,
to make sure that the signatory knew what he was doing.244
The no-reliance clause complements the parol evidence rule by
defining which documents and alleged statements should not be consid-
ered by the court if a dispute were to arise. The process of negotiating
the clause itself could convey valuable information, since the argument
that the counterparty should not rely on X could immediately raise a red
flag. The no-reliance clause can reduce juridical risk. If allowing certain
evidence in would substantially increase litigation costs without increas-
ing the accuracy of the verdict, enforcement of the clause would benefit
both parties, ex ante. No-reliance clauses, if enforceable, can play a
meaningful role in controlling the costs of litigation. As then-Judge Alito
noted, if no-reliance clauses are not enforced,
[t]he danger is that a contracting party may accept additional
compensation for a risk that it has no intention of actually bear-
ing. This prevarication may amount to a fraud all its own . . . .
[T]he safer route is to leave parties that can protect themselves
to their own devices, enforcing the agreement they actually
fashion.245
Abry Partners provides an example of a contract that attempted to
control litigation costs with a no-reliance clause.246 It involved the $500
million sale of a portfolio company by one private equity firm to
another. 247 The dealheavily negotiated by sophisticated parties
included a number of representations and warranties as well as an
indemnification clause and escrow account, which limited aggregate liab-
ility for all misrepresentations or breaches to $20 million, as the buyers
sole and exclusive remedy.248 In exchange for the indemnification clause,
244. Extra Equipamentos E Exportao Ltda. v. Case Corp., 541 F.3d 719, 724 (7th
Cir. 2008). The relevant clause reads:
Both parties represent and warrant that in making this Release they are relying
on their own judgment, belief and knowledge and the counsel of their attorneys
of choice. The parties are not relying on representations or statements made by
the other party or any person representing them except for the representations
and warranties expressed in this Release.
Id. at 730.
245. MBIA Ins. Corp. v. Royal Indem. Co., 426 F.3d 204, 218 (3d Cir. 2005).
246. 891 A.2d 1032.
247. Id. at 1035.
248. The indemnity for misrepresentation clause read as follows:
[T]he Selling Stockholder agrees that, after the Closing Date, the Acquiror and
the Company . . . shall be indemnified and held harmless by the Selling
Stockholder from and against, any and all claims, demands, suits, actions, causes
of actions, losses, costs, damages, liabilities and out-of-pocket expenses incurred
or paid, including reasonable attorneys fees, costs of investigation or settlement,
other professionals and experts fees, and court or arbitration costs but
the seller gave up all the materiality qualifiers of the representations and
warranties.249 The contract stated: [T]he provisions of [the indemni-
fication clause] were specifically bargained for and reflected in the
amounts payable to the Selling Stockholder . . . .250 The buyer sued and
asked for rescission; it claimed that some representations were inaccurate
and, had it known the truth, it would not have closed the deal.251 The
Delaware Chancery Court denied the sellers motion to dismiss.252 I do
not want to get involved with the merits of that decision. I simply want to
underscore the notion that sophisticated parties can price the buyers
reliance on the accuracy of representations, ex ante.
In the absence of an indemnification clause, the buyers reliance on
the accuracy of the representations and warranties is protected in two
ways. The accuracy at the time made and at closing is usually a condition
of closing. That is, if the representations were inaccurate, the buyer
would have an option to abandon.
The second form of protection is a money-damage remedy for
breach. The representations and warranties can survive closing for a
contractually defined period of time. If after closing the buyer learned
that a warranty was inaccurate, it could sue for damages. The buyers reli-
ance on the accuracy is protected, in part, by monetary damages. That is
simple enough. The more interesting question arises if the buyer learned
before closing, but still went through with the deal. Could it close the
deal and then sue for the breach? The answer in the key American juris-
dictions (New York and Delaware) is yes.253 Parties are, however, free to
contract over it. The practice has a namesandbaggingand it is
embodied in the ABA Model Stock Purchase Agreement:
The right to indemnification [or] reimbursement, or other
remedy based upon any such representation [or] warranty . . .
will not be affected by . . . any Knowledge acquired at any time,
whether before or after the execution and delivery of this
specifically excluding consequential damages, lost profits, indirect damages,

punitive damages and exemplary damages . . . to the extent such Damages . . .
have arisen out of or . . . have resulted from, in connection with, or by virtue of
the facts or circumstances (i) which constitute an inaccuracy, misrepresentation,
breach of, default in, or failure to perform any of the representations, warranties or
covenants given or made by the Company or the Selling Stockholder in this
Agreement.
Id. at 1044 (emphases added in Abry Partners).
249. Id.
250. Id.
251. Id. at 1065.
252. Id.
253. Charles Whitehead provides evidence on the use of sandbagging clauses broken
down by jurisdictions in which the default rule favors or opposes sandbagging. Charles K.
Whitehead, Sandbagging: Default Rules and Acquisition Agreements, 36 Del. J. Corp. L.
1081 (2011). The two most prominent commercial jurisdictionsNew York and
Delawareare pro-sandbagging. Id. at 1087.
Agreement or the Closing Date, with respect to the accuracy or

inaccuracy of . . . such representation [or] warranty . . . .254
In the absence of such a clause, some courts have held that the buyer
would have to prove reliance on the misrepresentation and, if the buyer
knew the truth before closing, it could not have relied. The alternative
formulation notes that the buyer relied when it set the initial price; if it
had known that the representation was false, the contract price would
have been less. Both sides can plausibly invoke reliance in their favor.
The weight of authority, and practice, is with the pro-sandbagging side.255
If the value of the target had increased prior to closing, the buyer
might want to complete the deal even knowing that the representation
was false. If it could not pursue its claim, it would end up with less of a
bargain than if the representation were accurate or if the contract price
had reflected the actual state of affairs. Two other factors argue in favor
of sandbagging. First, there is a moral-hazard problem. If the value of the
target has gone up, the seller could recapture some of the increased
value by breaching a representation. Absent sandbagging, the buyer
would have to weigh the gains from closing against the alternative of not
closing and suing for the breach. Second, there is a costly litigation prob-
lem; if a buyer were to sue for breach of warranty in an anti-sandbagging
jurisdiction, the seller could allege that someone at the buyers firm had
been made aware of the misrepresentation before closing, and the truth
of that statement would be a fact question that likely would be enough to
survive summary judgment.
CONCLUSION
Reliance, I am sure, shows up in many other nooks and crannies of

contract law. My purpose here has been twofold. First, I want to illustrate
the wide range of contract questions impacted by reliance. But second, I
want to emphasize that the importance of the reliance concept does not
translate directly into contract doctrine. That one party relies on the
continued performance of the counterparty does not suggest how, if at
254 . Mergers & Acquisitions Comm., Am. Bar Assn, Model Stock Purchase
Agreement with Commentary 299 (2d ed. 2010). Some agreements have anti-sandbagging
clauses. For example:
Buyer has no knowledge of any facts or circumstances that would serve as the
basis for a claim by Buyer against Sellers based upon a breach of any of the
representations and warranties of Sellers contained in this Agreement [or
breach of any of Sellers covenants or agreements to be performed by any of
them at or prior to Closing]. Buyer shall be deemed to have waived in full any
breach of any Sellers representations and warranties [and any such covenants
and agreements] of which Buyer has knowledge at the Closing.
Id. at 301.
255. Id. at 5556. Whitehead looked at all corporate acquisitions in a five-year period
in which the representations and warranties survived and found that very few had anti-
sandbagging clauses. Whitehead, supra note 253, at 109293.
all, that reliance should be protected. Doctrine in many areas does not
serve to facilitate contracting; rather it can be an obstacle that
transactional lawyers must overcome. To the extent that the default rules
are couched in moralistic terms (good faith, preventing injustice, etc.),
the transactional lawyers task is made more difficult.
In their Introduction, Fuller and Perdue emphasize the importance
of recognizing purpose:
We are still all too willing to embrace the conceit that it is possi-
ble to manipulate legal concepts without the orientation which
comes from the simple inquiry: toward what end is this activity
directed? Nietzsches observation, that the most common
stupidity consists in forgetting what one is trying to do, retains a
discomforting relevance to legal science.256
They were not concerned with the purpose of the parties themselves;
their focus was on damage rules that could be imposed on parties. Here,
the end to which the activity is directed is the end of the parties (sophisti-
cated parties, to be sure), not the doctrinalists. The parties end is to
balance the reliance against other factorsadaptation to change, juridi-
cal risk, allocating responsibility for controlling costs, and so forth. How,
if at all, they would protect their reliance is a matter of contract design.
The insights from considering the design problem should, in turn, influ-
ence the development of doctrine.
256. Fuller & Perdue, Reliance 1, supra note 1, at 52.


Protecting Reliance - Victor P. Goldberg PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Protecting Reliance - Victor P. Goldberg PDF

Hochgeladen von

Copyright:

Verfügbare Formate

GOLDBERG, Victor P. "Protecting Reliance". En: Columbia Law Review, Vol. 114, No. 4 (MAY 2014), pp.

Reliance plays a central role in contract law and scholarship. One

of the most cited articles in contract law. 4 Saying that reliance is

Because contracting parties need to adapt their relationship as circum-

12. (1854) 156 Eng. Rep. 145 (Ex.); 9 Ex. 341.

Traynors famous decision in Drennan v. Star Paving Co.,17 which involved

I. RELIANCE AND FLEXIBILITY

A. The Option to Abandon

unfortunate. When framed as a design problem, it becomes a matter of

interest to proceed. Delay or abandonment hurts it. Second, if the VCs

24. See Friedrich Kessler, Automobile Dealer Franchises: Vertical Integration by

over the wisdom of making long term commitments and long

26. 116 F.2d at 678 (quoting Bill of Particulars).

it was, read as follows: Kellogg generally encourages its employees to

He suggested that there were two types of enforceable preliminary

41. Id. at 498.

In the leading New York case finding a Type II agreement, Brown v.

rezoning), either party wanted to go it alone, it could trigger a buy-sell

56. Goldberg, Option to Unbundle, supra note 48, at 56.

in Hexion Specialty Chemicals, Inc. v. Huntsman Corp.62 provides a good

62. 965 A.2d 715 (Del. Ch. 2008).

75. Id. at 16.

them to stipulate that the promisor will be allowed to terminate

80. Id. at 23.

rhetorical power. If a breaching party is perceived as having wronged the

compensation to a universally applicable norm results more from mistaken path

If the measure of damages provided in subsection (1) is inadequate to

88. U.C.C. 2-708(2) (2012).

An interesting illustration is provided by Rodriguez v. Learjet, Inc.99

99. 946 P.2d 1010 (Kan. Ct. App. 1997).

Long-term variable-quantity contracts are quite common. They are

subcontractors and investments in capital. None of the OEMs cover suppliers

also a single-use, multimillion-dollar item.113 The primary use for the

there can be scenarios in which the guaranteed payments would exceed

123. 769 F.2d 1284, 128687 (7th Cir. 1985).

II. RELIANCE AND CONSEQUENTIAL DAMAGES

The role of reliance in a consequential damage claim is different. In

147. Goldberg, Lake River, supra note 135, at 439.

Contract doctrine classifies these consequential damages as expectation

151. Hadley, 156 Eng. Rep. at 145; 9 Ex. at 341.

spare parts. There is little or no reason to believe either that the

159. See supra Part I.A.

the scheduled departures or arrivals of the vessel or other conveyance

breach of warranty (to the consumer) or a product recall. Fords stand-

I must reiterate that these remain exceptions to the basic point.

III. RELIANCE AND RESTITUTION WHEN PERFORMANCE IS EXCUSED

Performance of a contract could be excused under various doc-

174 . Much of the following discussion is based on Victor P. Goldberg, After

In the first Anglo-American case recognizing the impossibility

started production. 186 Unfortunately, the Germans decided to invade

186. Fibrosa Spolka Akcyjna, [1943] A.C. at 34.

compensated after performance was excused; or it could simply leave the

IV. RELIANCE AND REVOCATION

The interplay of reliance and injustice shows up in other sections of

would swallow up the consideration doctrine has not come to passquite

The GCs reliance, therefore, is a question of fact. How could the

213. Id. at 76061.

218. Drennan, 333 P.2d at 760.

Drennan rule really gets it backwards.224 Instead of having reliance deter-

V. RELIANCE AND INFORMATION

of the seller. Reliance is implicated in both problems. Information is

were a general and vague merger clause, it would be ineffective to

if someone who is not a big boyindeed is not even represented